I will refer to the throw up, or barf, as “puke”.

Supplies

Here are the supplies I keep on hand at all times in case of stomach bug emergency.

1. Rubber doctor gloves — https://a.co/d/5WU5lhw
2. Clorox wipes — https://a.co/d/do8lH8K
3. Lysol Disinfectant Spray — https://a.co/d/6FWtOGc
4. Masks — https://a.co/d/fZbaBhd
5. A significant amount of hand soap. — https://a.co/d/gCHeoPv
6. Extra small garbage bags — https://a.co/d/bk3NITQ

The First Sign

Someone pukes. It usually comes out of nowhere. That’s how it begins. Here are your immediate next steps:

• Step 1: Remove all non-puking children from the puke area
• Step 2: Get rid of the puke. If there is time, put on some rubber gloves and a mask. Most of the time (for a child) the first puke is unexpected, and is not landing in a trash can or puke bucket. One parent should sacrifice themselves as the puke parent. This parent will clean all the actual physical puke from this point on. The other parent stands guard so no other kids get into the puke area. Blankets, clothes, or anything washable needs to go into the washing machine IMMEDIATELY and you start washing it. Ideally with bleach. Anything that you throw away, goes into a garage bag and walked right out of the house into the outside garbage.
• Step 3: DONT TOUCH YOUR FACE. Stomach bugs spread via particles of barf on your hands or surfaces. So the entire time you are interacting with puke or your puke child, do not touch your face.
• Step 4: Console your puke child. This usually means new clothes, new blanket, brush teeth, a puke bucket, a tv show, and a cup of crushed ice. NOTE: Put a clean garbage bag in the puke bucket each time. Don’t just let them puke into the bucket if you can help it.
• Step 5: Clean any hard surfaces that got puke on them with a disposable Clorox wipe. Ideally you are also taking the dirty wipes to the outdoor garbage.
• Step 6: WASH YOUR HANDS. This cannot be emphasized enough. Hand sanitizer does not work on stomach bug germs, so you HAVE to be washing your hands with a lot of soap and hot water. Wash your hands after every single time you touch your puke child or something in their sick nest. By the end of this, your hands should be raw from all the scrubbing. If puke got directly onto your skin at any time, wash that area twice.
• Step 7: Open the windows in the puke area. “But Shannon!” you say “it’s winter!”. I don’t care!!! We need the fresh air people, open them up. This would also be a good time to turn on an air purifier if you have one.
• Step 8: Spray everything down with Lysol disinfectant spray. And I mean EVERYTHING. Puke child’s room, light switches, stair railings, door knobs, toys, toilets, toilet flushers, faucets. Every room in your entire house. It should smell like febreeze in there. Spray every morning and night, and spray after a puke if it happens in a new location.

Ground Zero

Alright, so the first puke is done but now you know you’re in for it. In my personal experience the active puke window is 8 hours long from first puke. So for the next 8 hours the puke parent needs to sit with the puke child and wait for subsequent pukes. It’s going to be something like hourly. After each puke, tie off the garbage bag of puke and bring it outside. If they puke into something that doesnt have a bag, flush the puke down the toilet. NOTE: Close the lid of the toilet seat before you flush the puke so that the puke particles don’t go flying up and around your bathroom.

After every interaction with your puke child, wash your hands. Wash your hands BEFORE you touch your phone. If you forget, and you touch a phone or tablet or tv remote or something, then wipe the device off with a clorox wipe. Don’t give your puke child anything to eat or drink during this 8 hour period. If they really want something, ice chips.

Now is a great time to put in a grocery order for things like: saltines, pedialyte, gatorade, etc. Also buy anything that YOU would want in the event that YOU get the stomach bug, like ginger ale or whatever.

Now is also a great time to wash puke child’s bed sheets if that option is available to you.

Odds are, you will make it to the end of the 8 hours without multiple people puking. Then just monitor them for about 24 hours to make sure they are feeling better, getting some fluids, etc.

Another one bites the dust

Its the next day. The sun is shining and then you hear it: “Mom I dont feel good!” says a different child. You grab the puke bucket away from puke kid #1 just in time to catch puke from puke kid #2. If you’ve made it this far with only one parent being exposed to the puke, great job! This puke parent needs to continue with the next sick child. Repeat ALL above steps. Ideally, you have a “sick” area of the house, and a “not sick” area of the house, and the kids in those two areas never interact with each other during the puke windows.

Are we out of the woods yet?

3 days has gone by since the last kid has puked. “Huzzah!” you think “The nightmare is over!”. You would be WRONG! In my experience, you are not in the clear until the 8th day since last puke. YES THAT IS CORRECT. You cannot celebrate until the morning of the 8th day. This also means that you should probably not hang out with anyone for a week after having a stomach bug. This gets a little tricky with work and school, but I definitely wouldn’t see any newborn babies or grandparents during this time. Also, make special note that its 8 days since the LAST puke. So you if you had one kid sick on day 1, and another kid sick on day 3, then you need to wait until day 11 to see the elderly.

Day 6: The puke parent has fallen

It is the morning of day 6 and you are feeling great, when suddenly you are not feeling so great, and you need to go lie down. Soon enough, you have the stomach bug. The rules for adults are a little bit different because I assume they are self sufficient enough to deal with their own puke, and also able to wear a mask without constant nagging. The sick parent should isolate themselves for the next 8 hours in one room, ideally a bathroom. Nobody goes in. If you need to come out, you wear a mask. Special Note for new moms: You can breastfeed this entire time if you want to, just wash your hands before you touch your baby and wear a mask. When I personally am sick, I wait until the 8 hours is over to interact with the baby because its easier to keep myself clean.

After the first 8 hours, you can pick a comfier isolation room like a bedroom. Still wear a mask if you are traveling within the house and not in your isolation room. Make sure all the puke has been removed from the isolation bathroom, and spray it with lots of Lysol before moving on to another room. Text requests for toast to the non-puke parent and they will leave them by the door. Stay in the new isolation room and away from people for 24 hours. If someone tries to tell you to help with bedtime, remind them you have just spent the last 8 hours puking, and they can shove it.

Looks like we’ve made it

Its day 14 and no one else is sick. Hooray you’ve done it! You survived a stomach bug. And maybe you even learned something. I’m excited to do it all with you again in about 6 months…. good luck!

Step 1: Do you qualify for FMLA?

When you have a child (or adopt a child) you may be eligible for unpaid leave under the federal Family Medical Leave Act (FMLA). FMLA is a federal program, meaning anyone in the United States (male or female) can use this leave. It is 12 weeks of unpaid leave. (You can only take this leave once in a 12 month period. Typically, this is for a “rolling” 12-month period measured backward from the date of any FMLA usage.)

To qualify, you must meet all of these requirements:

• The employee must have been employed with the company for 12 months in the past 7 years.
• The employee must have worked at least 1,250 hours during the 12 months prior to the start of FMLA leave
• The employer is one who employs 50 or more employees within a 75-mile radius of the worksite

This is the backbone of your maternity/parental leave. Short term disability (which is covered in the next section) will run concurrently to FMLA time off. This means you can’t use 12 weeks of unpaid time, and then 6 weeks of short term disability for a total of 18 weeks off - you have to use them at the same time.

Another important thing to remember about FMLA is that it offers you job protection. This means you cannot get fired while you are out on FMLA. This is important because Short Term Disability does not offer job protection.

Step 2: Tell your employer

Now is typically a good time to tell your employer that you are pregnant. I have done this between 13 and 20 weeks. The legal requirement for the amount of time you need to tell your employer before you are out is 30 days:

Employees generally must request leave 30 days in advance when the need for leave is foreseeable. When the need for leave is foreseeable less than 30 days in advance or is unforeseeable, employees must provide notice as soon as possible and practicable under the circumstances.

If you are planning to adopt and you are placed with a child without 30 days notice, you would fall under the 2nd category of “possible and practicable”.

Typically, you want to tell your supervisor first, then your Human Resources (HR) representative, and then other co-workers. Usually HR will have some kind of a form to fill out that is just for the company saying that you will be requesting leave. If your employer has any kind of maternity leave program, now is good time to ask about that as well. HR will also give you information on Short Term Disability, because New York State requires employers to provide it (see next section).

Step 3: File a Short Term Disability Claim

Maternity leave typically falls partially under Short Term Disability (STD) insurance.

New York is one of the few states that has a temporary disability insurance program, which requires employers to provide short-term disability insurance for their employees…Pregnancy and childbirth are covered under the law.

While the above quote is accurate, DBL-exempt employers do not need to provide STD insurance. If you work for a DBL-exempt employer (a public school, a charity, etc.) then you may want to look into purchasing STD insurance privately. If you plan on purchasing STD insurance, be sure to do it before trying to get pregnant. If you are already pregnant and apply for STD, they will not cover your current pregnancy, but may cover future pregnancies.

The idea is that literally the physical act of giving birth disables you from working and so mothers get paid time off via STD insurance. Every STD plan is different. Contact your STD insurance company for more information. A typical amount of time off is 6 weeks for a vaginal birth and 8 weeks for a C section, paid at about 50%. For an adoption, you would not be eligible for STD.

Typically, you want to file a claim before you give birth. I worked with Liberty Mutual (or Lincoln Financial Group) because this is the plan my employer offers. You can do this by calling the company or going online. You usually answer some questions or fill out a form. You won’t hear anything more until after you would like to start your STD leave.

I did not experience this first hand, but I believe if you are put on bed rest by a doctor you can get additional short term disability leave via a 2nd claim. Also bed rest would be covered by FMLA.

An employee who becomes disabled more than four to six weeks prior to giving birth or continues to be unable to work more than four to six weeks after giving birth may have to submit additional medical documentation to support her claim.

HOWEVER, as mentioned above, STD does not offer job protection. Therefore, if you use 2 weeks of STD before your baby is born, you will only have your job protected for 10 weeks after they are born. After that, your employer may be able to fire you for excessive absence.

Step 4: Apply for New York State Paid Family Leave

In 2018 New York State (NYS) started a Paid Family Leave (PFL) program that will increase each year until 2021. Not many states have benefits like this, so it is nice that NYS offers it. This leave is available to both mothers and fathers and does not require you to physically birth the child (unlike STD).

To qualify, you must:

• Have worked for your employer the last 26 weeks, 20 hours or more per week OR
• Have worked for your employer a total of 175 days, with less than 20 hours per week
• Be employed by a non-DBL-exempt employer.

The last bullet means that typically an employer who is exempt from providing DBL (Statutory short-term Disability Insurance) will not provide PFL benefits either. These employers do have the option to voluntary opt-in if they so choose, but typically they do not. For more information check out the link above, but this means public school teachers, those who work for non-profits, etc. do not get to use NYS PFL.

In 2020, this program allows for 50 days of leave paid at 60%. (Check here for more information on the financials.) NYS PFL does not run concurrently with STD. Meaning if you qualify for STD then you can take all of that leave, and then start the 50 PFL days after that. Another great thing about the NYS PFL program is that it does offer job protection! So even if the 12 weeks of FMLA runs out, your job and health insurance are still protected.

Applying for NYS PFL happens generally through an insurance carrier that manages the PFL claims for the company you work for. So in my case, this was Libery Mutual again. Here are the generic steps from ny.gov:

1) Employee notifies employer 30 days prior to leave, when practical.

2) Employee fills out a claim form according to employer instructions. Claim forms are available from employer, insurance carrier, or ny.gov/paidfamilyleave.

3) Employee obtains supporting documentation for leave (birth certificate, military deployment certification, etc.)

4) Employee submits claim form and supporting documentation to insurance carrier or as directed by employer. Insurance carrier must pay or deny a claim within 18 days of receipt of the completed claim.

This is the claim form provided by NYS. You might use this form, or you might use a different form created by the insurance company handling your PFL claim. Typically, you will want to state your intent to use PFL days before your child is born, but it may be hard to apply because it can sometimes require knowing the date of birth and the exact days you will be taking leave.

Step 5: What does your employer offer? How does that interact with these other government policies?

Every company is different. Some may offer no parental leave, others may offer 6 months or more. Ask your HR representative what your company offers. There may be rules such as how long you have to have worked for the company, and what percentage of your pay you will receive. You will receive whatever is the highest amount of money between your company policy and the government policies, but not both. You will not make more than 100% of your paycheck while out on maternity leave.

For example, let’s say your employer offers 12 weeks of 100% paid maternity leave. STD or PFL would run concurrently with your employers program, and your employer would be allowed to accept the money that you would receive from these government programs on your behalf. So if you were to receive 60% of your paycheck on STD during the first 8 (or 6) weeks of maternity leave, then your employer will actually receive that 60%, and then give you that money plus an additional 40% in your paycheck from them. It will work the same way for NYS PFL. If your employer plans on playing you 100% of your paycheck for the 4–6 weeks after STD ends, they will be able to collect PFL checks on your behalf and use that money to pay your salary. You will still get your full paycheck from your employer, but you will also be using up PFL days at the same time. Here is a quote from ny.gov:

If the employer has an internal policy that provides the employee with full wages while the employee is on Paid Family Leave, the employer can pay the employee up front and then request reimbursement from the insurance carrier, up to the amount of the Paid Family Leave benefit.

If, such as in the timeline above, your employer does not offer as much paid time off as is offered from the NYS PFL program, then you can use the remainder of your PFL days after your employer parental leave program has ended. So in the example showed above, the employee would have 30 days of PFL left to use after the 12 weeks of paid leave has ended.

Step 6: Plan your leave

If you qualify for FMLA, STD, and PFL, you will get a total of 80–90 days (depending on C-section or vaginal birth) of partially paid maternity leave which is 16–18 weeks. You can either take this time all at once when the baby is born, or split it up. Typically leave under FMLA isn’t intermittent (but it does not have to be the full 12 week duration). Your employer can agree to allow leave to be used intermittently, but they aren’t required to allow it. STD needs to be taken all at once after your child is born. You cannot split the 6 weeks of STD by working in between. PFL can be used all at once or in 1 day increments. This is great because you can do something like work 3 days a week while you are adjusting back to full time work. You cannot take half PFL days; it only works for full day increments. Obviously, you can always use sick time or vacation time to supplement these policies. NYS does not have any laws against “use-it-or-lose-it” vacation policies (like some other states) so it may be hard to save up a lot of time this way, but it is something to consider when planning your maternity leave.

Step 7: After the baby is born

Once the baby is born, you will have to call/email all interested parties to update your leave plan. This will include your employer, the company managing your STD, and the company managing your PFL. You may also need to provide some kind of “proof” such as a birth certificate.

After you give birth you will call the STD insurance company to report which day the baby was born (not always your due date) and which day you started STD (not always the baby’s birth date). The STD insurance company will either issue checks directly to you, or to your employer if they have monetary parental leave benefits that are higher than the amount you would receive with STD. The same goes for PFL payments. See the explanation of this above in Step 5.

You probably also have to finishing filling out your PFL claim form and send that in. Be sure to do it before or within 30 days after you want to take your first PFL day. If you are taking leave continuously (meaning all at once) you will only need to provide a start and end date. If you are taking days intermittently, you will need to individually list out each of the 50 dates you plan to take PFL. The checks that you receive for PFL do not have taxes taken out of them. You can make federal and state tax payments during the year to avoid penalties at the end of the year when filing taxes.

Step 8: Take and report your leave

To return to work you may need your doctor to fill out a work release form and return it to your employer. It is a good idea to bring this to your 6 week postpartum visit. You will probably need to check in with your supervisor occasionally (maybe once a month) while on leave to confirm that your plans have stayed the same (or report a change). The STD company may call you when your leave is finishing up. If you are using PFL days on an intermittent or somewhat intermittent schedule, you may need to call more frequently or fax time cards to the insurance company managing your PFL claim for each day that you use. Don’t rely on your employer, the insurance company, or the government, to keep track of the days you take off and pay you accordingly. During my first stint of maternity leave I was both overpaid and underpaid at different times. Its important to keep a record of all the days you took off and all the checks you received during your time on leave. Occasionally, time cards and paperwork will get “lost in the mail” and you will have to file again.

In the end, I ended up splitting time off with my husband giving my children 6 months at home before daycare. While you will want to maximize the amount of time you spend at home with your child, the amount of income you can afford to lose will also be a big factor into your decision making. It’s a personal decision and you might have different feelings about it after your child is born than you did beforehand. I hope in the future that every parent has the option to take as much time as she feels she needs without pay being a concern.

Reasons I dislike Perforce:

I have been using Perforce for over a year. It is a terrible, legacy version control system, which I hate. Haters will say “iT’s beCAUse yoU DONT UnDerstaND it!” but I will counter that with “No one understands it, and if I - a smart human being - can use it for over a year and still have complaints, then clearly I am not the problem here”.

1 ) You have to check out files and add new files separately. There is no button to do all of that at the same time. This becomes particularly annoying when you are moving files. The files at the old location get deleted while the new files are never added back in. Or worse, the new files get added and the old ones don’t get deleted so they just float around as a second older version of the same file.

2) There is no way to just see all of the changes that have been made to the workspace in one list. Many times I add a file and forget its even there, and then my changelist is not complete when I try and get it reviewed.

3) Explicit Checkout. You have to check out a file before you can edit it, so Visual Studio is constantly popping up warning messages. If I start editing a file, it should be an automatic check out. This would be so much more convenient.

4) Having different workspaces that all have access to the same depots in the “Depot Tree” is confusing. There should either be only one workspace, or you should have to specifically designate which depots you want to be a part of that workspace.

5) What is a stream? If a stream is supposed to be a branch, then what is a change list? Streams only serve the purpose of giving me more things to maintain. Supposedly a stream is supposed to solve the problem of letting you know which depots to have in your workspace. However, I can still see all the depots, I just get errors like this when I choose the wrong ones to check out.

6) Why do your changes go away in your local workspace when you shelve files? This makes it hard/impossible to show others your progress while still working on changes, because Shelved files cannot be edited. There is a magical check box that let’s you get rid of this “feature”. But it should never be checked by default, who would want this?!?!

7) It is not a popular tool and is not taught in colleges. Nobody knows how to use it so getting new employees/interns to use it correctly is a giant pain.

8) When I click “Get Latest” (in P4V) and I’m on the Workspace view, I only want to get the latest of the folders I have already downloaded and put in my workspace. Instead, perforce tries to start downloading a whole bunch of other stuff that I did not actually intend to download. The “Get Latest” command should only download new things when you are on the Depot view of P4V, not the Workspace view.

9) Exclusive locks on files. It makes me sad that exclusively locking a file is even possible. It’s like all of Perforce is built because you don’t trust the other developers trying to work together on a project. Exclusive locks are applied automatically when you check out a binary file. You can only lock files that you have checked out, so you need to have them sitting in a changelist even though you clearly don’t want anyone editing them. What’s worse is, you can lock a file that someone else already has checked out! So they can be in the middle of working and you just lock it like “sorry no can do”.

10) Why does the concept of a “shelved” file that you cannot edit even exist. So you can share your changelist? I feel like Perforce is pretty obviously designed around not wanting to have two people edit a single file at the same time, so why would you even care about this? I’m constantly having to shelve things for review, then submit the actual files, and delete the old shelved versions. WHY. why.

11) You cannot submit shelved files from a changelist that also includes non-shelved files. Why allow changelists to include both types of files if they cannot be submitted together?

12) When using P4V, the log icons do not align with the log. So they end up just being a very confusing line of icons.

13) Perforce is a high-maintenance tool. Because it’s centralized SCM, every operation touches the server. In order to make that performant, they keep a lot of working memory — a journal — unpersisted or persisted in swap space. So any kind of server maintenance is complex and time-consuming

The fact that GitFusion even exists should tell you that I am not alone in preferring Git over Perforce. Here is a great stack overflow article that explains some of these pain points for a person who is more familiar with Git.

Swarm is a code review tool built by Perforce and it’s also awful

Swarm is really the only code review tool built to work with Perforce (because no one uses Perforce) so you’re pretty much forced to use it if you want to do code reviews. It is missing some pretty basic features and has some other strange behavior which I list below:

1 ) Everything you click on in swarm is super delayed. (Opening files, opening reviews, etc.)

2) You can only comment on a singular line of code. There should be an option to select multiple lines or a block of code to comment on.

3) You cannot delete comments on a review. So you end up writing “EDIT: nevermind” at the bottle of your comments. smh.

4) When you are on a review page, you have the choice of viewing changes in line and side by side. Most of the time I want to view changes side by side, but occasionally I want to view them in line. There are two places you can toggle this information on the review page. One is at the top of the page, and the other is at the top of each individual file. If I click the button at the top of the page, it should change how I am viewing every single file, but instead, it only changes the files I have open, so I have to go through and click the button on the top of every file anyway. The madness!

5) You cannot change your profile picture on Swarm. You have to use a randomly selected avatar or make a Gravatar account. This makes it so every single user has one of 3 randomly selected bee themed avatars. If your team all “randomly” ends up with the same avatar, you can’t tell easily who is making which comments.

6) Reviews are not associated with groups, just people are. So when you want to see all the reviews in a group, you are actually seeing all of the reviews any person in that group is a part of, even if it is for a different group.

7) To mitigate the issue of reviews not being associated with groups, I tried to use a project. Reviews are associated with projects, but they are not adjustable. You can only associate a review with a project by writing #review in the changelist description of the files you are shelving, and then the review is generated under the project associated with that depot or file. Once a review has been started, you cannot associate it with a project.

8) Review names are just numbers, nothing useful (example: review 383854)

9) It is not easy to see changes for updates to a code review. Is this from the first change set, or the last change set? The default setting for a review is showing the base version with the latest version, even if you have already looked at the code review and commented. Swarm should be smart enough to know that I have already been here and then change the default to seeing only the updates since my last comment.

10) Automatic email notifications from Swarm do not always get sent, so you have to manually send regular emails to people that you want to review your code and when you’ve made updates.

11) Reviews don’t get “closed”, they get “archived”. Confusing nomenclature. It is also very unclear what happens when a review gets “Approved”. What if it is approved by one person but not another? Does this close the review and prevent updates?

12) Great errors like this: (This happened because I tried to add a user as a reviewer but had the wrong username. A better error would definitely be “user does not exist”).

Thanks for making it to the end of my long list of grievances. My team is now using mostly git so I am a happier person than when I first wrote this. Woohoo! Would love to hear from anyone that has found a way to get around these problems.

Some Background

At the ripe old age of 22, my husband and I packed up our two cats and moved across the country to Silicon Valley. It was an awesome experience and thankfully we had an employer footing the bill for our move. We kept track of every moving related transaction and have documented it here for your enjoyment.

Moving your possessions

To move our possessions we used U-boxes from U-Haul. We have actually used U-boxes for moving twice now and it was in general a pretty good experience. There are several different “pod” style moving companies, but we found U-Haul to be the cheapest.

We were packing up a one bedroom apartment so we used 2 U-boxes. U-Haul allows you to book and have delivered one extra box than you think you will use (so in our case, 3) free of charge. This way if you accidentally run out of room you are not up a creek. But you do have to pay more to get that extra box shipped if you end up using it.

To get our two boxes shipped to CA, we paid ~$3,000 (with a $358.82 deposit). The price for shipping U-boxes changes every day on their website. I learned this the hard way. I recommend giving yourself plenty of time and checking the website pricing everyday for a week for the same trip. Then you can continue to check after that until you see a price that is lower or around the lowest price you saw in the first week. They will not honor older prices if you did not book (even if you have a screenshot. Speaking from experience. Lol.)

Once all our items were packed in boxes in our apartment, we carried the boxes and furniture from our apartment into the U-Boxes ourselves. This was a mistake, but we had a first floor apartment so we thought we could do it. I would definitely hire movers for the loading, it is not too expensive. We did hire movers for moving things into our California apartment because we were on the 2nd floor and there was no elevator. We hired our movers through U-Haul’s Moving Help and were pleased with the results. I would highly recommend getting professionals to move your stuff. They are way better at fitting things into the boxes to maximize your space, and they are way faster than you and 15 of your friends. We hired two guys for two hours. They did amazing. You also have to buy your own locks for the U-Boxes while they travel. We got Master Locks; something like this amazon link.

Moving your bodies

We flew Delta from Rochester (ROC) to San Jose (SJC). We also had to check some extra baggage and pay for 2 cats to fly with us in the plane under the seat in front of us. The cat carriers actually cost about $30 each but we had a gift card. We got trip insurance because I was nervous about the flight, but I wouldn’t recommend it. There is very little that is actually covered by a travel insurance policy except for like unexpectedly being arrested or so sick you cannot travel. Don’t forget the airport snacks!

Moving your vehicles

We shipped our cars! The car shipping industry is pretty under developed but make sure you get a quote and reservation early to avoid being overcharged. This takes about 2 weeks for them to drive your car across the country. (They don’t actually put miles on your car, it is loaded onto a big truck). You can also put things in your trunk. We shipped both of the cars we owned, which in hindsight was a mistake. You barely need 1 car to live in a city, let alone two. Coming from an area with a much smaller public transportation system, I could not fathom living without a car for both of us.

Preparing your new home

We were moving into a new apartment and we had gotten rid of a lot of furniture before moving. So we spent A LOT at IKEA getting our new apartment prepared. This is completely up to you how much you spend here.

DMV Fees

When you arrive in CA or your new state of residence, you will need new drivers licenses, license plates, and car registrations. Unfortunately the above is all I have documented in terms of paid transactions for this part of the move. We probably paid some with cash or check. I looked up online the cost of a Class C California license and it is $36. I also looked up the cost of registering a new car in Santa Clara county today, and that is below. So that brings us to a total of $514, but its possible things were cheaper a few years ago when we actually moved.

Forwarding your mail

You can sign up for mail forwarding with the US Postal Service online for $1 per name. You can sign up for free if you actually go in to a post office but who does that.

Taxes

You used to be able to deduct moving expenses from your federal taxes, but that is no longer the case as of 2019. However, if you or a member of your household is in the military you can still do this.

Total

In total, these amounts come to $7,022.15. We moved in the summer months, which may have helped cut down our costs. If you count the amount we spent on IKEA furniture, the total is $8,977.54. If I was doing it again, I would sell both cars before moving and save myself ~$2,500.

Some Background

At the ripe old age of 24, my husband and I packed up our two cats and moved across the country to Rochester, NY (our hometown). Thankfully we had an employer footing the bill for our move. We kept track of every moving related transaction and have documented it here for your enjoyment.

Moving your possessions

To move our possessions we used U-boxes from U-Haul. We have actually used U-boxes for moving twice now and it was in general a pretty good experience. There are several different “pod” style moving companies, but we found U-Haul to be the cheapest.

We paid $65.79 for packing supplies (empty boxes).

We were packing up a one bedroom apartment so we used 2 U-boxes. U-Haul allows you to have one extra box reserved (so in our case, 3) free of charge. This way if you accidentally run out of room, you are not up a creek. But you do have to pay to get that extra box shipped if you end up using it.

To get our two boxes shipped to NY, we paid ~$3,300 (with a $383.61 deposit). The price for shipping U-boxes changes every day on their website. I learned this the hard way. I recommend giving yourself plenty of time to check the website pricing everyday for a week. Then you can continue to check each day after that until you see a price that is lower or around the lowest price you saw in the first week. They will not honor old prices if you did not book (even if you have a screenshot. Speaking from experience. Lol.)

Once we had all our boxes packed, we hired movers to carry the boxes and furniture from our apartment to the U-Boxes. We hired our movers through U-Haul’s Moving Help and were pleased with the results. I would highly recommend getting professionals to move your stuff. They are way better at fitting things into the boxes to maximize your space, and they are way faster than you and 15 of your friends. We hired two guys for two hours on both ends of the trip (loading and unloading the boxes into our new house). They did amazing both times.

Storing your possessions

We had to keep our U-Boxes in storage for a little over 2 months while we waited for our house to close. This is what we were charged to store them. You also have to buy your own locks for the U-Boxes while they are shipped and stored. We got Master Locks; something like this amazon link.

Moving your bodies

We flew Delta from San Jose (SJC) to Rochester (ROC). We also had to check some extra baggage and pay for 2 cats to fly with us in the plane under the seat in front of us. Don’t forget that Lyft ride to the airport!

Moving your vehicles

We shipped our car! The car shipping industry is pretty under developed but make sure you get a quote and reservation early to avoid being overcharged. This takes about 2 weeks for them to drive your car across the country. (They don’t actually put miles on your car, it is loaded onto a big truck). You can also put things in your trunk. We moved our bikes this way.

Preparing your new home

We had our house carpet cleaned and regular cleaned before moving in. You may have other expenses such as replacing an electrical box or buying new furniture. We also bought several smoke detectors not listed here.

DMV Fees

When you arrive in NY or your new state of residence, you will need new drivers licenses, license plates, and car registrations.

Forwarding your mail

You can sign up for mail forwarding with the US Postal Service online for $1 per name. You can sign up for free if you actually go in to a post office but who does that.

Taxes

You used to be able to deduct moving expenses from your federal taxes, but that is no longer the case as of 2019. However, if you or a member of your household is in the military you can still do this.

Total

In total, these amounts come to $7,972.49. We moved in the summer months, which may have helped cut down our costs.

1. Hire Software Engineers who are passionate about quality to do the testing
2. Require code reviews for tests
3. Have Quality Engineers review production code
4. Write tests for your tests
5. Put time into your test framework and get developers involved in enhancing it
6. Focus on the testing pyramid.
7. But make sure you are not lacking End to End coverage
8. Run your tests every night, and with every merge (or push)
9. Monitor pass and fail rates of each test

Hire Software Engineers who are passionate about quality to do the testing:

The number one thing you can do to help your test automation succeed is to make sure you have the right people/person leading up the initiative. I would highly suggest looking for these 3 things when hiring a Quality Engineer:

Someone who can code

Don’t hire someone who cannot code. If you hire someone who is not technical, they will have a hard time writing automation, participating in code reviews, creating or utilizing a DevOps strategy, and many other things you will see in this list. These things are essential to the overall success of a test automation framework and you need to hire someone qualified to do the job, plain and simple.

Someone who is passionate about quality

Don’t hire someone who really wants to be a developer. You want to hire someone who is passionate about quality and will love the work they are doing. Someone who will stay in the position long term, and who is not just looking at it as a stepping stone to becoming a developer. This is key to the long term success of your framework.

Someone who is influential

Don’t hire someone who cannot enforce new, forward-thinking quality policies. Hiring someone who is influential is important because introducing a new test automation framework can sometimes come with a lot of push back. This new Quality Engineer will probably be walking into an established team that already does things a certain way. Asking a developer to review the results of a test suite before every merge to master may seem like a lot of new work that they did not sign up for. Make sure you have hired someone who can stand their ground and explain why certain quality policies will provide benefits to everyone in the long run.

That is not to say that it will not be difficult. You will probably have a difficult time finding someone who is both able and willing to do this job. One reason for this is cultural stigma. There is a very prevalent and widespread philosophy among Software Engineers that only people who “can’t hack it” as a developer become a tester. Only people who are less skilled, less technical, or less smart, would ever voluntarily choose to do the job of a tester over that of a developer. And that is simply not the case. You will find people who have struggled to find a development job, or who are fresh out of school, who will apply to a testing position out of desperation. That is not the kind of person you are looking for to fill this role.

Require code reviews for tests:

Code reviews have been proven again and again to be well worth the effort you put into them. This is no different for test code. Just because it is not customer facing, doesn’t mean it doesn’t need to be reviewed. Reviewing code decreases its likelihood for bugs, which is a character we want in our automated tests. We want automated tests that are accurate and reliable so that when they fail, they are revealing true issues and not flakiness in the test framework. Developers need to review the test code as well as Quality Engineers. They need to be reviewed for quality of test cases (Are we testing the right things? Should any additional cases be explored?), as well as structure of the automated test code. (Is this code readable, reliable, and maintainable? Pretty much anything you would ask in a regular code review you could ask in the test code review as well.)

Have Quality Engineers Review Production Code:

In a traditional setting, software quality assurance testers are usually expected to be less technical than their developer counterparts. They may or may not work along side the developers on the same team. They may or may not be actively involved with the decisions made about the product during development of a feature. Today, this should no longer be true. Best practices highly suggest including testers on the agile product teams through all parts of the software development life cycle. This includes code reviews. A Quality Engineer should be assigned to every code review the team puts out. This creates a number of benefits for the team and the engineers.

Firstly, Quality Engineers get the opportunity to point out issues early on. Aside from requirements review, you can’t shift much farther left than a code review. This helps cut down on bugs that will make it through to the next phase in the life cycle. (If you are thinking “but my quality engineers will not be able to catch any bugs in my production code, they are not technical enough or familiar enough” then you have a bigger problem and please refer to tip #1.)

Secondly, doing code reviews on production code gives Quality Engineers an opportunity to learn more about the product they are testing. In depth knowledge of a product is essential for good testing. Quality Engineers should be the power users of the product. They should know the system inside and out and they should know the edge cases. Any insight that you can give them is beneficial. You may now be thinking “but what about black box testing? Isn’t that important too?” I have two responses to this. The first response is black box testing isn’t nearly as useful as you think it is, and there really isn’t anything that a black box tester is going to catch that a good white box tester can’t catch. The second response is, it is my opinion that black box testing is only useful in a user experience testing environment. Black box testing is supposed to be done by a user that knows nothing about the system’s internal structures or workings. If you are trying to do usability testing on your UI, I would absolutely do black box testing. But I don’t think having a Quality Engineer do UI/UX testing is a good use of resources. Use alpha and beta releases for that. Get a true customer perspective.

Finally, code reviews allow Quality Engineers to learn about good coding practices. Everyone can benefit from reviewing another person’s code and seeing how they do things. Maybe they will see an interesting pattern they haven’t seen used before. Or maybe they will see another developer commenting on a bad coding practice that they weren’t aware of. Knowledge share is a great tool, and having Quality Engineers and developers do code reviews together strengthens your team and their technical prowess.

Write Tests for your Tests

Write unit tests for your test framework libraries that are used often. Let’s say you have a very complex part of your framework that randomly assigns a user particular permissions and privileges, and then runs tests using that user to ensure that the permissions are upheld in multiple scenarios. How will you know that these tests continue to work as you expect over time? This is where unit tests come in to play. You can use unit tests to make sure the right set (or a key subset) of privileges are getting generated each time.

Write tests for data validation. Make sure the data passed to your test suite “makes sense”. Using tests for data validation will help ensure that you are giving the test framework what it needs to succeed. Nothing is worse than finally convincing a developer to kick off an automated test run, only to realize they’ve supplied the wrong arguments and they will be disappointed in the results. Developers are not typically people who give grace in the department of a test framework. They want something as easy to use as possible. Any kind of overhead and they will lose interest. This is one way to set your test automation framework up for success. [Here’s an example: Chances are, your automated test suite has a global configuration file. Something that is either being altered by the user before it runs, or something manipulated by a CI framework like Jenkins. Either way, this is going to be the place for input into your test framework. Let’s say one of these variables is a server address that you want the tests to run against. Does this server address need to be an IP address? Does it need to have a port? Does it need to have a specific protocol? (http vs. https?) Create a test that runs first, and determines that the correct format argument is supplied here. Then you are ensured that the wrong address will not be the reason your tests fail.]

Put time into your test framework and get developers involved in enhancing it

Nothing is for free. The team needs to put in time to get the best out of an automation framework. In general, investing in a good test automation framework will (at first) require more time than some manual testing. However, you will quickly see the fruits of your labor in subsequent releases. The team won’t need to go back and invest heavily in manual regression, because their automated tests are doing the work for them. Its important that everyone on the team have a basic understanding of the test framework and how to write tests. Pair programming is helpful with this. Developers are developers for a reason. They love creative, out of the box solutions. Use their strengths to help enhance your test code. If they were going to create an architecture document for the test framework, what would be the outcome? What design patterns would they implement? Put their ideas into the framework and you will see not only better quality tests, but more buy-in from the whole team.

When asked to quantify the benefits of a test framework, its hard to actually identify specific defects that the framework is preventing (because we’re catching them so early, that we don’t need to write them!) This can sometimes make it difficult to convince your manager or your team lead that the time you put into a test framework is actually being fruitful. Test automation may pay off over a long amount of time, but it does pay off. The more often your team releases, the greater the payout will be. Here are a couple of case studies [1] [2] to help make your case.

Focus on the testing pyramid.

The testing pyramid is a pretty common resource among Quality Engineers. It can be helpful to keep it at the forefront of your mind, and show it to developers. The basic idea is that a team should focus the highest quantity of tests on unit tests, followed by service tests (or subcutaneous tests), and finally UI tests. This is because the longer a test is, the harder it is to maintain. There are a lot of resources on the internet about the testing pyramid so I won’t go into it too much more here.

But make sure you are not lacking End to End coverage

A test suite is not complete without at least some level of end to end testing. You need to know that the whole system can interact together, minimally through your most common work flows. You don’t want to wait until the end of a release to realize that someone changed a service and they didn’t realize you relied on it, and now your workflow is broken. If you have a few key scenarios running every night or with every push, you will catch these issues. It can sometimes be a problem that a team is so afraid of inflating the amount of UI or end to end tests that they produce that they skip them altogether. I found an interesting quote below explaining why the testing pyramid may not always be the bible:

Don’t get me wrong, in 2019 the testing pyramid, TDD and unit tests are still a powerful technique and are probably the best match for many applications. Only like any other model, despite its usefulness, it must be wrong sometimes. For example, consider an IOT application that ingests many events into a message-bus like Kafka/RabbitMQ, which then flow into some data-warehouse and are eventually queried by some analytics UI. Should we really spend 50% of our testing budget on writing unit tests for an application that is integration-centric and has almost no logic? As the diversity of application types increase (bots, crypto, Alexa-skills) greater are the chances to find scenarios where the testing pyramid is not the best match.

One of the issues I have seen with teams that skip automated End to End or UI testing is with performance. Automated tests (typically) push your system a little more than a manual tester when it comes to utilizing your UI and your whole system. A workflow that works well when one or two people are using a system may completely break when 20 different automated tests are running at the same time and testing different things. (This is not to say that you shouldn’t have dedicated performance testing, you absolutely should. I have rarely seen environments where performance testing is run on a nightly basis, but if you are doing that — good for you. But I’m not really going to get too much into nonfunctional testing here.)

Run your tests every night, and with every merge (or push)

Continuous integration is a huge component of automated testing. Any automated tests that are not run on a regular basis are going to get stale quickly. Issues are not caught as soon as they could be when tests are only run once a week or once a month. This is especially true if a lot of test alterations need to be done before the monthly automation run can take place. Like anything, small investments over time yield larger benefits and are a lot easier for someone to tolerate. (Would you rather run a mile every day, or 7 miles every Saturday?)

Don’t rely on a person to manually hit “run”. This is going to require some DevOps knowledge. Get familiar with Jenkins or another continuous integration tool. Work on getting a staging environment that your tests can run against each day. This will help a lot with your efforts to shift left. Ensure that the results of every test run are available to the whole team to monitor. Make sure everyone is aware when something breaks. Ideally, enforce a rule that no one can merge (and especially not release) without a 100% passing rate. All test failures must be investigated, ideally by the developer initiating the merge.

Monitor pass and fail rates of each test

When you start running your tests on a more regular basis, you are going to start noticing tests that fail intermittently. What gives?! We call these tests “flaky tests” and they are normal to happen a small amount (about 1.5%) even at the best companies out there (even Google!). There are many reasons for this that I won’t get into now, but the question remains: how do you make sure your team maintains its faith in the test suite? The answer to that question is this: closely monitoring the failure rate of every test to determine if it is still a good candidate to be in the release test suite. One way of doing this is to implement an Inspector class into your automation framework which logs each test result in a database. Then set up a monitoring dashboard that aggregates that data into an easy to follow “leader board” of flaky tests. Here is an example of what that dashboard could look like for your team:

Establish a goal pass_rate for each test; for example 70%. These results include every retry. Monitoring this dashboard can be a manual process done by the team. When a test falls below the passing rate, remove it from the release suite and create a ticket for it to get some love and attention. This will work well when working with a smaller suite of tests, but it’s easy to make the jump to an automated system that will remove these tests automatically from the release test suite and create the defect tickets all on its own. This system of removing a test from the release suite is called a “Demotion”. I have had great success with maintaining the integrity and reputation of a test suite among developers using this method. When a test fails during a deploy, it is investigated for real failures by the developers.

I hope these 9 tips will help you on your journey to creating and maintaining a reliable software testing framework. Has something else outside of these tips worked really well for you? Let me know in the comments!

INTRODUCTION:

The history of the development of computer science (CS) curriculum began at Purdue University in 1962 when the first CS department was established [12]. It began very theoretical and has branched out over time. In the 1980s computer science departments around the country saw a large influx of applicants as CS became more popular and well known. Rochester Institute of Technology (RIT) had the first bachelors of software engineering program in the United States [13]. These introductory programs were the beginning of structured teaching of these subjects. Before degree programs, most of computer science was something that a person would teach themselves out of necessity. As these degree programs became more prevalent, teaching styles varied from university to university. One of these variations, which has been a topic of discussion fairly recently, is to include agile techniques into introductory CS education. There are many different possible ways to include agile practices into CS education. Some of the more popular methods used include the practices of Scrum and Extreme Programming (XP), the two most popular agile methodologies [9]. In this paper we will be specifically examining, Test Driven Development, Pair Programming, Daily Standups, and a few others. RIT includes some aspects of these practices into their curriculum. These include the try system, partnered projects, and many others. These practices can be improved by more closely adhering to agile practices as they would be used in a business setting.

Retention rates for CS classes across the country are much lower in comparison to other programs. The annual attrition rate among freshmen and sophomores majoring in CS in the U.S. has been reported to average 19%, and at some schools to be 66% [6] Some of this can be explained by general attrition in US colleges and universities [7], but in comparison to other fields, STEM and Computer Science programs have a lower retention rate. This problem stems from a lack of appropriate introductory classes. Referred to by some as “weed out” classes, these difficult starter classes have the strong tendency to deter students from continuing in a field that they fear will only increase in difficulty.

“Attempting to weed out entrants is a disservice to the whole of computer science: its teachers, students and practitioners. Weeding out entrants has a noble intent to help students discover that computer science isn’t for them at an early stage, however it is insular and conceited. Many — perhaps most — entrants will do well to jump directly into the deep end of the computer science pool. However room should be allowed for those who want to splash around in the shallow end first.” [8]

Assimilating agile practices into early computer science education would have the goal of making the material easier to understand. If this was possible, then retention rates would rise and fear of difficult or impossible to comprehend coursework would diminish. With that in mind, we write this paper as an observation of both the advantages and disadvantages of agile practices being used in early education, as well as a discussion on the most appropriate means of including them.

WHAT IS AGILE?

Agile is defined by Oxford’s dictionary as

“relating to or denoting a method of project management, used especially for software development, that is characterized by the division of tasks into short phases of work and frequent reassessment and adaptation of plans.”

Traditional engineering approaches were failing programmers, and they needed to rethink their idea of software development. Agile practices and methodologies were created in response to a pattern of failed software projects. All of the agile practices focus on promoting the four main values of agile:

• individuals and interactions over processes and tools
• working software over comprehensive documentation
• customer collaboration over contract negotiation
• responding to change over following a plan.

Hints of these principles can be found throughout all of the agile methodologies. Each of these methodologies have their own sets of practices and ways of doing things, but over time they have overlapped and are generalized as simply Agile.

CONTEXTUAL OVERVIEW:

Here we present a brief overview on popular agile practices used in many of the popular agile methodologies including Scrum, Extreme Programming, and others. Understanding the main principles upon which these practices are based will help the reader to understand their role in early computer science education, as well as their importance to professional software development.

Test Driven Development:

The main concept of test driven development (TDD) is to write all of the tests for a program (or particular module of that program) before the actual development begins. This has many advantages; there is always a suite of tests ready to go (initially and at a refactoring stage), and there are never parts of functionality that are forgotten about and not implemented. TDD was first used primarily in Extreme Programming (XP), but has gained a reputation as a useful technique amongst other methodologies as well, and is not used only in XP.

Pair Programming:

Pair programming is the practice of having two people work at one computer while coding. The theory behind this is that having more people looking at the code and actively participating in its development will result in better code. This has been proved in several cases. One example is a study done by Laurie Williams at the University of Utah [11]. In this study, it is proved that the correctness of code increases by 15% on average, with a 20–40% decrease in the amount of time it would have taken to write the code as an individual. These are both stunning advantages, deterred only by the fact that you would be paying two salaries instead of one. Pair programming is used mostly in XP where it was first introduced.

Daily Standups:

A daily stand up is a meeting with a software development team where each attendee aims to answer three questions:

1. What did you do yesterday?
2. What will you do today?
3. What impediments are standing in your way?

These questions allow for the full team along with the project manager to get an appropriate outlook on the team’s status. These meetings are meant to be kept as short as possible as to not take up too much of the day with bureaucracy. The daily standup is a concept introduced in scrum that is now used by both software development teams and other engineering teams.

Iterative development:

Iterative development is a software development methodology that focuses on repeated releases, increasing functionality and complexity in each release. Iterative development means that the product is built in iterations, where each iteration expands on the software product until the project is completed. The process is also incremental, that is, each iteration includes parts of all phases of the traditional software development model. Each iteration contains analysis, design, code development, and finally testing. Iterative development is a concept that is used frequently in agile software teams today.

TEACHING COMPUTER SCIENCE:

Traditional teaching style of CS

Computer science classes traditionally follow a very Waterfall-like approach. The Waterfall Method is a software methodology which strongly resembles methodologies in other engineering fields. In this process, developers will go through a rigid structure of requirements gathering, design, implementation, verification, and maintenance [14]. Every step must be fully completed before moving on to the next, and none can be revisited once it has been left. This kind of software development is seen often in introductory education. A professor will assign a task or set of tasks for the students (requirements). Then the students may (or may not) briefly create their plan of action for the assignment (design). Coding becomes the majority of the time spent on the project (implementation), with testing occurring if there is time at the end (verification). Maintenance is not a step in this scenario since the project is simply handed in and not used over a large period of time. This is the scenario played out in almost all computer science classrooms taught, with mixed results and success rates.

TEACHING AGILE:

Agile is currently being used as a means of teaching in some institutions in the United States. There is also a lot of theoretical literature on what may or may not be effective for students.

How does RIT teach Agile?

RIT is one of the leading institutions for software engineering. While not directly using agile methodologies to teach computer science, traces of agile principles can be seen in a number of activities. The first is the Try system. The Try system is a set of UNIX programs developed at RIT. It is used for electronic submission of assignments and projects. The Try system allows professors to develop a suite of tests to run on the submitted files. This allows students to get immediate feedback on their assignments. This is representative of the Test Driven Development style of software development in that students will often write code to pass the tests. Professors could truly use TDD in the classroom setting if the Try system was expanded to give students more robust feedback on their programming. One thing that would be necessary if RIT chose to truly incorporate TDD into introductory education is the modification of the Try system to run programs with randomly chosen values. This would prevent students from simply hard-coding values in their program in order to pass the Try tests. RIT could also modify the Try system to give students the ability to add their own tests to the system, gradually increasing the effectiveness and breadth of the test suite over time.

RIT also uses partners for a large number of projects. This is somewhat representative of pair programming. It is not true pair programming however as that requires both members of the team to be working at the same computer. One of the advantages of pair programming in an introductory setting is the shared learning that can occur between team members. If one member is weaker than the other, the stronger member can share their knowledge to help teach the student. One concern however is the weaker student being carried by the stronger student. If the stronger student simply does all the work, then there were no gains from assigning the assignment as a pair programming task. Pair programming would be hard to implement for homework assignments for this reason, but should definitely be investigated in a lab type setting.

Calvin College Case Study [1] [2]

One specific case study we looked at was a compiler class at Calvin College. Originally this class was taught on a traditional semester schedule using traditional methods. Students would start small and build up their compiler to more and more complexity as the semester progressed. At the end of the fifteen week schedule students would be evaluated on whatever final state their compiler was in at that time. This system was increasingly difficult for students. Most of the time, students would get about ten weeks in before the parts that they were trying to add to their compilers became too complex and difficult. Students would spend so much time trying to get full functionality on these difficult sections that they never actually got to any assignments after that. In the end, most students would have a lot of knowledge on what happened at the beginning of the semester and virtually no knowledge on what was taught at the end. The subject of this case study was to see how the addition of agile methodologies and practices would make the class easier for students. The class changed its approach from delivering a fully functional and complex product at the end, to building small functioning compilers over the course of the class, and adding layers of complexity as the semester continued. This iterative development, along with TDD practices allowed for students to gain a greater understanding of the compilers. The case study was such a success that Calvin College was able to use the curriculum developed as a class taught over interim, or 15 days between semesters. Even if it was difficult for students to grasp some of the minute complexities of the compilers, they would at least be able to grasp the overarching concepts of each part of the compiler during the first few iterations.

How can various Agile techniques be used?

One of the techniques used in agile development is test driven development. TDD is a useful concept to introduce early on as it teaches students the importance of testing, which is often a subject that students do not spend much time on. Most colleges do not offer a separate major for software testing. While testing is touched upon in early classes, it is often not until later in the student’s academic career that they are exposed to testing in detail. Testing software is one of the most important parts of the development process, and so entwining testing and development in introductory computer science could prove of great benefit to computer science students. One way test driven development helps is that it allows the professor to create robust test suites for assignments. They can then provide their test suites to the students. Students will then be able to use these test suites as a guidebook in their development, testing each method they implement as they develop. The test suites will act as robust, detailed outlines for the assignments, allowing the professor to limit hints on the actual implementation of the assignment.

Pair programming is another technique that holds value when teaching computer science. Some of the benefits of pair programming are significantly less bugs, overall code quality, and shared knowledge between developers. One of the biggest risks involved with using pair programming in education is having one member of the team take over. If one member is significantly stronger than the other, then they might just take over and complete the assignment by themselves, defeating the purpose of pair programming. This will harm the weaker team member significantly as they will not actually get to learn from the assignment. It is very important for the stronger member to assist the weakest member, not to complete the assignment for them. On the other hand, if the stronger member helps guide the weaker member through the assignment, then the lesson is reinforced for both of them. To account for this, pair programming should be used primarily in a lab setting. This will force the two members of the pair to work on the same workstation - one of the most important aspects of pair programming. A teacher will also be able to monitor the pairs to ensure that they are contributing equally and assist pairs with problems that they get stuck on.

Daily standups are another agile practice that can add to the classroom setting. Daily standups enhance team spirit, which help help get students involved with the success of the class. Another notable benefit of using daily stand ups in the classroom setting is that it encourages transparency. Computer science can be very difficult at times, especially in the early classes when there are still fundamental concepts that must be learned. Hearing from your peers about the roadblocks that they are having or things that they did to complete the previous day’s assignment could greatly enhance the morale of struggling students who feel that everyone else just ‘gets it’. Sharing their status every day will also get students comfortable with working on teams. When properly adopted to the classroom setting, daily standups could prove very useful to decrease attrition rates in computer science education.

ADVANTAGES AND DISADVANTAGES:

There are many varying opinions on the best way to teach computer science to students. Many researchers who have taught and explored this topic for years have established well respected opinions on this topic with significant proof. Here we compile a list of both the advantages and disadvantages sited in these arguments for or against agile techniques being used in early computer science education.

What are the benefits of introducing agile techniques early on?

Agile practices can have some really beneficial effects on students and the learning process when implemented properly. A lot of the advantages refer to the practical uses of knowing how to use agile in their careers later on. TDD technologies like JUnit are worth while to learn and will come in handy in a practical sense. Learning those practices early on really ingrains it in your experience and it will become second nature [1]. TDD is also a useful practice for refactoring. When a test fails, the developer can tell that something was broken even if they weren’t intentionally trying to change that functionality. It’s important that students know how to test software upon graduation because it is something that they will have to do their entire career. The traditional waterfall approach of projects makes testing the last activity that happens. students have a tendency to procrastinate resulting in the testing portion of almost every waterfall style project getting short changed because of time constraints. If this were to consistently happen across the board it would lead to students who were clearly lacking in testing skills in comparison to their other software development skills [10].

Pair programming can be beneficial when used properly. It brings objectivity to coding between pairs. Pair programming encourages transitive learning between students, helping reinforce concepts taught in class. It is also a very useful agile technique to know as a new graduate instead of encountering it for the first time after graduation. Pair programming is a useful technique in the agile toolbox, so understanding how to use it early could allow students to start using it to solve complex problems later on in their education.

An advantage to using a spiral technique for teaching is that the students are shown the same material multiple times (each time being slightly more in depth) [1]. This will allow teachers to get more and more detailed with complex topics, to ensure that students are proficient in each topic by the end of the class.

What are the disadvantages of introducing agile techniques early on?

While learning to use agile can be very worthwhile in a software engineering career, there are some disadvantages to coming upon it as a novice developer. Many agile practices rely strongly on the individual and their ability to follow the agile procedures. New students may not have the discipline or dedication to do that [1]. Throwing a student into an environment that uses agile practices that they are unfamiliar with as well as expecting them to learn course content is a dangerous mix. Chances are in an effort to focus solely on course content, the agile practices will suffer. This would be unfortunate for many reasons. Many can attest to the fact that agile practices when not followed properly can be just as dangerous (if not more) than not using them at all. Another downside to using agile practices on a group of students who aren’t ready would be the possibility of them forming bad habits. Bad habits would be extremely hard to break later in their careers.

Another disadvantage to using agile techniques in introductory education comes about when estimating time to completion and planning for deadlines. Planning projects in an agile way means the developer needs to have the ability to estimate coding time. This is difficult for all programmers, and even more so for inexperienced ones [1]. Students who don’t really have any past projects to base their estimates off of would have a difficult time planning sprints or increments for development that would match up accurately with class deadlines.

Finally, for classes hoping to utilize TDD, the knowledge of how to efficiently write tests and use TDD software would need to be taught. Class time will be spent teaching the class how to use the TDD technologies [1] instead of focusing on the course material on its own. Curricula would need to be adjusted to include time for these extra technologies to be learned, and instructors would have to be able to adjust their focus from the main content of the class in order for the students to be successful for the majority of the year.

These disadvantages definitely need to be taken into consideration and mitigated before deciding to implement agile practices fully into a university’s curriculum or even a few specific courses.

RELATED WORK:

A paper by David Parsons at Massey University demonstrates the possibility of teaching agile through games [4]. These games were created to coach students on the basic concepts of different agile practices. Not only are games enticing, but they do a good job of showing students the benefits of using agile throughout their programming careers. Playing one of these games at the beginning of the course would give them more of a willingness to participate in agile throughout the semester. One of the major disadvantages of using agile practices is the unwillingness of students to put forth their full effort into promoting the strategies. If these games showed them the importance and overall concept of the practices they were meant to be emulating for the duration of the course, students would have a better appreciation for the good that would come out of them and would work harder to use them.

We decided to take one of these games and play it in real life to see how relatable it was to agile. The game we chose was The Pen Game. In this game, students would be required to pair up similar to how they would in a pair programming environment. Each pair is given the necessary pieces to assemble a pen. One student would be the driver and the other would be the navigator. The driver is the only person allowed to touch the pen pieces, while the navigator gives him instruction on how to properly assemble it. We played this game with a sample size of about 20 students, resulting in 10 pairs. Results varied amongst the pairs, but in the end we would consider the game a success as all teams were able to successfully complete the activity while following the instructions. Students enjoyed playing, and understood in the end the importance of listening to your partner to determine what works and what doesn’t.

OUR CONCLUSIONS:

The benefits and practical nature of agile methodologies allow it to be used effectively for a large number of purposes, and education can also benefit from them. Applied correctly, agile methodologies can decrease the retention rates of computer science education. Additionally, seventy one percent of software organizations [15] currently utilize agile methodologies. The widespread adaptation of agile methodology in the software industry means that it should be introduced early on in education as well. Early computer science classes should introduce agile techniques such as test driven development and pair programming. This will help prepare students for more complex systems that will require robust testing, and get them prepared for group work. Once the fundamentals of computer science have been learned, assignments should largely shift to a group-based approach. Larger scale systems should be assigned to teams of students who will implement requirements using an agile approach. Heavy guidance from teachers throughout the agile process will allow students to learn how to be a member of a successful agile team. Teaching with agile is not without downsides. Test driven development is a very specific way of developing code, and will have to be taught to the students before a class that uses it. An introductory class on software testing, taken concurrently with the student’s first programming class, would help alleviate this downside. Pair programming also has risks of lopsided pairs. This could be dealt with through surveys and exercises to identify the relative strengths of the students, and then pairs assigned based on that. Finally, this practice would ensure that students know the benefits of using agile practices and that they are as invested in their success as the instructor is. This would work to mitigate the lack of discipline that students may have, as agile relies on it’s developers to have good discipline.

FUTURE WORK:

Including agile practices into early computer science education is a very broad topic that could be expanded in many ways. It would be really interesting to compare retention rates from a university that used agile techniques in education and ones that did not and to see if that actually makes a difference, assuming that the students, professors, and class content were comparable. One could also observe different combinations of agile practices in an attempt to determine which worked best and resulted in the best student understanding. It would be a difficult feat to accurately determine what worked best; some ideas could be standardized testing or a general student survey for understanding of course content.

SOURCES:

[1] Frens, Jeremy D., and Andrew Meneely. “Fifteen compilers in fifteen days.”ACM SIGCSE Bulletin 38.1 (2006): 92–96.

[2] Williams, Laurie, and Robert Kessler. Pair programming illuminated. Addison-Wesley Longman Publishing Co., Inc., 2002.

[3] “Lab 7b: Advanced Control Structures.” Processing & Java: An Introduction to Computing. Calvin College, 3201 Burton St SE, Grand Rapids, MI 49546. 17 Mar. 2014. Web.

[4] Parsons, David. “Creating game-like activities in agile software engineering education.”

[5] Kamat, Venkatesh. “Agile Practices in Higher Education: A Case Study.”SlideShare. Goa University, 23 Feb. 2012. Web. 14 May 2014.

[6] Sloan, Robert H., and Patrick Troy. “CS 0.5: a better approach to introductory computer science for majors.” ACM SIGCSE Bulletin. Vol. 40. No. 1. ACM, 2008.

[7] Lotkowski, Veronica A., Steven B. Robbins, and Richard J. Noeth. “The Role of Academic and Non-Academic Factors in Improving College Retention. ACT Policy Report.” American College Testing ACT Inc (2004).

[8] Rentzsch. “CS Is Not for Me.” Tumblr. N.p., 12 Feb. 2011. Web. 11 May 2014.

[9] “7th Annual State of Agile Development Survey.” Version One, 2013. Web. 26 Apr. 2014. <http://yogi.se.rit.edu/~swen-356/resources/7th-Annual-State-of-Agile-Development-Survey.pdf>.

[10] Bergin, Joseph, et al. “Agile development in computer science education: practices and prognosis.” ACM SIGCSE Bulletin 37.1 (2005): 130–131.

[11] Laurie Williams, Robert R. Kessler, Ward Cunningham, Ron Jeffries. “Strengthening the Case for Pair Programming.” IEEE Software. IEEE Software, July — Aug. 2000. Web. 4 October 2013

[12] Rice, John R., and Saul Rosen. “History of the Department of Computer Sciences at Purdue University.” Computer Sciences History. Purdue University, n.d. Web. 14 May 2014.

[13] Lutz, Michael. “Professor Lutz on the History of Software Engineering at RIT.”YouTube. YouTube, 18 Apr. 2014. Web. 14 May 2014.

[14] “Requirements Analysis.” Cs.tufts.edu. Tufts University, 16 Sept. 2009. Web. 14 May 2014.

[15] “Success Rates Rise — Tranforming the high cost of low performance”. Project Management Institute. Web. 2017.

Categories and Subject Descriptors.

D.2.9 [Software Engineering]: Management — productivity, programming teams, software process models. K.3.2. [Computers and Education]: Computer and Information. Science Education — computer science education, curriculum, information systems education. K.6.3 [Management of. Computing and Information Systems]: Software Management — software development, software process.

General Terms

Human, Factors, Management, Standardization

Keywords:

Agile, Pair Programming, TDD (Test Driven Development), Scrum, Extreme Programming, Spiral, Waterfall, Daily Standup

ACKNOWLEDGEMENTS

Thanks to Professor Malachowsky for teaching this class.

The prices associated with hand-me-down items are how much I estimated those items to be worth.

“No” means “not a gift, I bought this myself”.

“Did not get” means “although I originally thought we would need this, I never ended up buying it or receiving it as a gift.

Clothing

Most of the clothing for our baby we were given as gifts or hand-me-downs. When we were 12 weeks pregnant we bought some Stanford University baby clothes. It will be hard for you to make it through your whole pregnancy without buying at least SOME splurge clothing.

Not listed here but it should be is SLEEP SACK! After your baby can’t be swaddled anymore they need a sleep sack to sleep warmly because they cannot have blankets. We did not buy any sleep sacks while I was pregnant but bought 2 after baby was born, both around $25.

Feeding

If you are breastfeeding you may still want to get an emergency can of formula. I bought a small 90 oz can for $17. I didn’t use the pumping bra until I was back at work. I already had a bottle brush & bin to clean dishes in. If you don’t have these you will want them as well.

Accessories

We got 2 mirrors and 4 sun shades because we have two cars.

Health and Baby Care

Nursery

We got the bassinet from a thrift store. In hindsight this was a bad idea because you NEED baby’s bed to be safe. Buy this new and you will sleep better at night. Also we should have gotten fitted sheets for the pack and play (which is listed in accessories but baby can sleep in as well.)

Bath & Diapering

Once you spend $1000 on your amazon registry you get $100 in diapers or other baby supplies for free. We had 1 box of newborn diapers, 1 box of size 1, and 1 box of size 2.

Daycare

This is what we paid to reserve our spot at Doodle Bugs.

Toys

We also got several hand-me-down books that are not valued here, and over the course of your baby’s life you will get more and more toys. These are just the toys and books we had before she was born.

Conceiving and Other

Normally your health insurance premiums are deducted from your paycheck each month to pay for your health insurance. If you are taking an unpaid leave of absence, you will be responsible for paying these premiums. If you are being paid, then the premiums will continue to deduct from your paychecks as normal.

Doctors Appointments

This is the amount that I was actually charged at each appointment. This will vary a lot depending on your health insurance but it gives you a good idea of what appointments to expect. For more information about that check out Every Pregnancy Doctor’s Appointment. I also had 3 different health insurance plans over the course of my pregnancy (highly do not recommend but what can you do) so the numbers here are all over the place.

Birth

I had a vaginal delivery and an epidural. The “Jeannie money” listed here is the amount of money that was charged to our baby as the patient, as opposed to the mother as the patient.

This will obviously be different for everyone, but what really got me was how much the hospital does not want you to know how much this is going to cost. Even the bills afterward are very hard to read, and I have no idea what I am actually paying for. Also, why are there so many separate bills? And why do they come months and months after the procedure?

Totals

Our total costs of having our first baby end up at $7,454.03. Of this, we paid $4,174.27, and we were given in gifts or hand me downs items that I valued at a total of $3,098.76.

• Weeks 4 to 28: 1 prenatal visit a month.
• Weeks 28 to 36: 1 prenatal visit every 2 weeks.
• Weeks 36 to 40: 1 prenatal visit every week.

But anyone who has had a baby will tell you that this is not all that is required! And if you are anything like me, you would appreciate a lot more detail. Here is a list of every medical appointment I went to while pregnant with my first child. This won’t be everyone’s experience, and we did have some complications that led to more appointments than a “normal” pregnancy. We also moved from California to New York at 20 weeks pregnant which added some appointments. But this is a comprehensive guide to each appointment and what to expect:

6 weeks: Phone appointment

We were trying to get pregnant, so we found out we were pregnant right at 4 weeks (or one day before my missed period). I was worried about a chemical pregnancy and waited one week before calling the doctor to schedule an initial appointment. When I called the office I was scheduled for a phone appointment with a nurse the following week. Before that phone appointment I was asked to fill out an online form about my pregnancy history and the pregnancy history of my family. While on the phone we reviewed that history and scheduled the first in office appointment for 8 weeks. They also gave me an emergency after hours number to call in case I started bleeding before that appointment.

8 weeks: First appointment

At our first in office appointment they confirmed the pregnancy via vaginal ultrasound! We were able to get a picture of the tiny, tiny baby and hear the heart beat. Some doctors will perform a pap smear at this time and test you for STDs; mine did not. After this appointment we went to a counselor’s office where we were presented with our options for genetic testing and we got to choose what kinds of tests we would like to proceed with. Because I was 24 years old when we got pregnant and had no previous pregnancy history, I was not offered testing such as an amniocentesis. All of the tests we were offered were blood tests and we chose to do all of them.

8 weeks: Blood Draw

At 8 weeks was the first time I got blood taken for pregnancy! You will get used to this quickly as they seem to need a lot of your blood. I went to a clinic instead of the actual doctor’s office. Here are some things you are tested for, as well as being a cystic fibrosis carrier (or a carrier of other diseases, depending on your race):

Your first prenatal visit will include blood tests to determine your blood type; your iron level to see if you’re anemic; your blood glucose level to check for diabetes; and your Rh factor (if your blood is Rh negative, and the father’s is Rh positive, the fetus may inherit the father’s Rh-positive blood, which could cause your body to make antibodies that would hurt your unborn child). You will also be tested for HIV, hepatitis B, and syphilis, as well as whether you’re immune to rubella (German measles), since coming down with this illness while pregnant, especially in the first three months, can cause birth defects.

12 weeks: blood draw

This blood is used for genetic testing that goes along with the ultrasound the next week (or around then).

13 weeks: Ultrasound

Time for another ultrasound! This is more what you would typically think of for ultrasounds (not vaginal). You will hear the heart beat. The size of the baby at this appointment will help determine your due date, especially if you have an abnormal period cycle. They will look at a number of things in this ultrasound including measuring the back of baby’s head for down syndrome markers (should you so choose).

16 weeks: blood draw again

At some point during your second trimester (for me it was at 16 weeks) you will go for another blood draw called the “Quad screening”. See more details below.

Your doctor will also take other blood tests (known as a triple screen or a quad screen) that will measure blood levels of alpha-fetoprotein, estriol, hCG (human chorionic gonadotropin), and inhibin, which can indicate whether the fetus is at risk for abnormalities such as Down syndrome or spina bifida.

17 weeks: Belly Check

This was my first (and only) belly check appointment in California. These appointments let you listen to the heart beat and ask any questions that have come up since your last appointment. It was helpful for me to create a list on my phone of questions I had throughout the last couple weeks so I remembered to ask them at my appointment. The results for my second trimester blood test came back one day after my scheduled 17 week appointment, so I was not able to review any of the test results at that time.

17 weeks: Meeting with a genetic counselor

The results came back for my second trimester blood test, and they showed a high level of alpha-fetoprotein or AFP. This could be indicative of your baby having spina bifida or other neural tube defects. A typical AFP score is between .5 and 2.5, and mine was 3.49. Because of this blood test we will go on to have several additional appointments needed for a high risk pregnancy. The first of those was this appointment with a genetic counselor. The state of California requires you meet with a genetic counselor when you receive (bad) blood test results like we did. The genetic counselor explains the test results to you, and what your choices are moving forward. This appointment was just in an office building and did not require any medical exams. We did a deep dive into my family history of miscarriage and birth defects, as well as my husband’s.

17 weeks: anatomy scan with a perinatologist (checking for neural tube defects)

A high AFP score is indicative of neural tube defects such as spina bifida. Because of this high score, we needed to have an early anatomy scan to look at the baby’s spine and head to see if any defects were present. Anatomy scans are typically done at 20 weeks pregnant and take about an hour. They ask you to drink a ridiculous amount of water before you come so that you start the appointment with a full bladder. Some things to know about this:

• The amount of water they ask you to drink is a GUIDELINE. You know when your bladder is full, and that’s all they need
• Your bladder needs to be full so that they can look at the uterus and other parts of your anatomy (not the baby).
• They like to do this at the beginning of the appointment but its not a requirement, so don’t feel bad if you don’t have to use the bathroom when you first start the ultrasound. Your bladder will fill up during the appointment and they can check your uterus at the end.
• Make sure you ask the ultrasound tech to let you know when your bladder doesn’t need to full anymore so you can go use the bathroom and then continue the appointment. That way you don’t lay there for the whole hour having to pee.
• If you are having your anatomy scan for an unfortunate reason like we were, it may be tempting to ask questions or try and read the face of the ultrasound tech to determine if your baby is “passing the test”. This is really hard to do and virtually pointless. A doctor or perinatologist will review the results with you right afterward anyway, so you might as well wait for them.

19 weeks: anatomy scan with a perinatologist — part 2

We “passed” our 17 week anatomy scan and were told there were no apparent neural tube defects for our baby. Woohoo! However, because we were only 17 weeks along, the doctor recommended a follow up ultrasound 2 weeks later. We did the exact same entire anatomy scan over again at 19 weeks with a different perinatologist. We had the same results (all clear), and also found out we were expecting a girl. This perinatologist suggested we have growth ultrasounds every 4 weeks starting at 30 weeks to ensure the baby was growing properly. Sometimes when pregnant women have unexplained high AFP scores (which was the case with me) their babies and placentas have growth problems. These growth ultrasounds would watch out for that scenario.

21 weeks: nurse visit; changed doctors

At 20 weeks pregnant we moved from California to New York! We had to find a new doctor, and this was our first appointment with that new office. Basically we just discussed the pregnancy so far and what else would happen during the 2nd trimester. This meeting was in an office and did not require any medical examination.

21 weeks: belly check and general wellness exam

This general wellness exam was required by my new doctor because I had not had a pap smear at the beginning of my pregnancy. Looking back I wish I had said no to this test because I was very confident I did not have any STDs, and I bled afterward because of the extra sensitivity of the cervix during pregnancy. But I was told this appointment was required before they would schedule the anatomy scan completion appointment for me.

21 weeks: anatomy scan completion

At 21 weeks we had our 3rd anatomy scan! However this one was not as abnormal. This scan was only a “completion” scan, to look at some areas of the baby’s heart that were not visible on the 19 week ultrasound because of the position the baby was in at the time of the appointment. Everything looked good. It is very common that a perfectly healthy baby will not be in the correct position for the usual 20 week anatomy scan, and you will need to schedule a follow up.

22 weeks: blood draw

At 22 weeks I had a crap ton of more blood drawn from me to do some standard/ “required” New York State testing. I was told we couldn’t reuse the results I had had from previous blood tests in California. What a bummer.

23 weeks: belly check

My first “normal” belly check appointment of the pregnancy. At each of these appointments I could ask questions and listen to the heart beat. They also typically felt my stomach for the position of the baby, and measured my belly with a measuring tape to see how I was progressing. I was also weighed, had blood pressure taken, and had to pee in a cup at every appointment.

24 weeks: leg pain appointment and ultrasound

If you have anything weird happen to you, you will probably call your doctor and they will bring you in for a not previously scheduled appointment. I had been experiencing pain in my left leg since about 20 weeks pregnant and had been worried it was due to a blood clot (because of the amount of time I had sat still while traveling over the last month). When the pain got particularly bad one evening I called the doctor who scheduled an appointment for me the next day. At that appointment they told me that what I was experiencing was probably sciatic nerve pain, but just in case they wanted to ultrasound my leg and check for a clot. I did the leg ultrasound and no clot was found. To mitigate my leg pain I bought a pregnancy belt which helps to lift the baby off of your sciatic nerve. It worked until about 35 weeks.

27 weeks: 1 hour glucose test

Sometime in your second trimester you will have to go in for the glucose test! I was able to do this test at the same clinic that took my blood, instead of going into the doctor’s office. This test requires you to drink a sugary koolaid like drink, wait one hour, and then get your blood drawn to test your blood sugar. If you pass this test, no follow ups are needed. If you fail, you have to do a follow-up 3 hour test. I passed.

28 weeks: Belly check

30 weeks: Belly check

Nothing particularly interesting happened at this appointment except that I noticed I had still only gained 12 pounds. I was assured this is totally normal and as long as the baby looks to be growing at the correct pace, it’s totally fine if moms don’t gain a ton of weight. 20–30 pounds is typical in a healthy pregnancy.

30 weeks: growth ultrasound

We had our first growth ultrasound! This did not take as long as the anatomy scan and only measured certain parts of the baby (as opposed to everything, like the anatomy scan). They will also measure how much amniotic fluid is present around the baby. Finally, you will get an estimated weight of the baby.

32 weeks: belly check AND 3rd trimester nurse visit

This was just a regular belly check with an added discussion about what my plans would be for the 3rd trimester and beyond! I also got a flu shot at this appointment (because it was winter) and a Tdap shot that everyone gets in their 3rd trimester of pregnancy to help protect the baby from whooping cough.

33 weeks: growth ultrasound

34 weeks: belly check

36 weeks: belly check

At this appointment they started checking my cervix for effacement and dilation. Typically the first time they check you is just for a baseline. I was 0 cm dilated at this appointment.

37 weeks: belly check

I was tested for Group B strep and was positive! For this you must take penicillin during labor. You have to take this every 4 hours during labor via an IV. Also if you are Group B strep positive you can’t labor at home after your water breaks, and you can’t have your “membranes stripped” to try and naturally induce labor.

38 weeks: growth ultrasound

We had our 3rd growth ultrasound and were told that I have too much amniotic fluid. This condition is called polyhydramnios and is a measurement called AFI or Amniotic fluid index. A normal AFI is between 5 and 24, and mine was 25. Because of our high AFP score, our doctor recommended seeing another perinatologist for an ultrasound at the end of the week.

38 weeks: perinatologist ultrasound and consultation

The baby was checked for blockages in the intestines, problems with the kidneys, inability to swallow, etc. No indications of any of that. Therefore the perinatologist did not recommend early induction.

38 weeks: belly check

High blood pressure! Barely, 130/80. Asked about pain walking, lightning crotch, and baby hiccups, all normal.

39 weeks: belly check

Blood pressure is lower now, pain walking is a little bit worse. Scheduled induction for 4 days past due!

40 weeks: Belly Check

Baby’s heart rate was 185 so I was put on the fetal monitoring machine for 40 minutes. Baby’s heart rate went back down to 150s and stayed there for the most part. All good. Couldn’t do membrane sweeping because I am group B strep positive. Sad

40 weeks: ultrasound to check fluid level and growth

I was still considered polyhydraminos at this appointment. The baby was also estimated at a weight lower than the growth ultrasound at 38 weeks. Because of these two things and the high AFP my induction was moved up 2 days to today!

Delivery!

Induction was very easy for me thankfully; I did not even need pitocin. My water broke about 12 hours after arriving at the hospital and the baby was born 9 hours after that. I got an epidural at 7 cm dilated about 3 hours before the baby was born. Baby’s heart rate was erratic throughout labor, and my placenta detached from the uterus wall before the baby was born. Because of those two reasons the doctor was worried we would have to do a c section. The baby had the umbilical cord around her neck twice so it had to be cut immediately. Delivered vaginally. Sounds scary but in the end she was a happy and healthy baby!

In the past six months I have been asked to be the matron of honor in two weddings! This has given me the opportunity to look at the wedding budget spreadsheet that I made four years ago when I was getting married. I poured over this thing for my entire engagement, and I kept it as accurate as possible. (I also use Mint.com, so it was easy to know exact amounts of what I spent without having a shoe box full of receipts.) Here I present to you, the only wedding budget spreadsheet you will ever need. You can view my estimations of what I thought my wedding would cost, and what I actually paid for everything. Make a copy of the google sheet for yourself and alter it based on your own expectations.

Here’s a couple of baseline assumptions that went into my estimations and ultimately the cost of my wedding:

• I wanted a pretty traditional wedding. Friends and family, dinner and dancing, the whole nine yards.
• I was getting married in Rochester, NY which has a pretty low cost of living. If you are getting married in NYC or SF this budget will be very hard to recreate (but you can still use the spreadsheet)
• I was 20 when I got engaged and 22 when I got married.
• We knew that other people would be paying for our rehearsal dinner, bachelor/bachelorette parties, and my bridal shower, so those events weren’t taken into consideration.

Choosing the date of your wedding:

When we were picking our wedding date, we knew we had a two year engagement ahead of us. Any venue I contacted had barely started booking two years out, so I was getting first pick. Because of this I was able to get the June 25th wedding date we wanted. But here is an interesting caveat that I did not expect — When you pick a very popular wedding date, other people you know are going to get married on the same day as you. We knew two other couples that got married on the same day as us. Honestly this didn’t really bother me, because you barely have time to talk to people on your wedding day, but something interesting to think about when you are choosing a date.

I did not want to get married on a Friday or a Sunday. If you do this, it will save you money because most venues will give you a discount price.

I did not want to get married in the winter, but there are a lot of vendors (not just including your venue) that will give you a discount for booking their services during an off month. The winters in Rochester, NY are very cold, and very susceptible to blizzards. We have a lot of out-of-town family and we wanted to make sure they were able to come to the wedding, so a winter wedding wasn’t an option for us.

The Reception

Your reception is what will use up most of your budget, so keeping your costs low here is really important. I looked at a few different places but finally ended up going with Glendoveers because of (it’s great location and food but also) it’s awesome prices. I was able to lock in my wedding price a year and half before my wedding at $34.95 a plate for a buffet. This really helped keep my costs down. They have since raised their prices slightly, but the quality of the food and venue were amazing, and their current price is still really decently priced for this area of NY. Also their wedding planner, Shannon, is really nice and helpful.

To open bar or not to open bar? Originally we had chosen to only have beer, wine, and soda for guests ($5.95 pp), because it was cheaper and neither of us are big drinkers. When it came down to it, we had a minimum cost for the event that we had to meet for our venue (and we had more people RSVP no than we expected) so we paid for the full open bar at $10 pp. Most packages I have seen come with a champagne toast for your guests without charging you, but you might run across one that is not free.

Something to keep in mind about minimums and the general cost of your reception — they do not include tax and gratuity. And usually “gratuity” is just a mandatory surcharge (you don’t get to decide what to tip). This is important to think about when budgeting and visiting venues. Ask about required gratuity percentages and know what you can afford after having paid 20% gratuity and 8% tax.

Splurge items from our reception:

There were a couple of things that we decided to splurge on in this category, and that helped contribute to us going about $1,500 over budget in the end. You can easily skip these items for your own wedding and it will help you save on costs:

• We rented gold chargers (plates) to dress up our tables. Depending on what kind of charger you want, you can buy them all outright for the same amount that it would cost to rent them, but then you need someone to be in charge of dropping them off, setting them up, and bringing them home. I opted to rent them and was happy with my decision. We did not need chair covers because the chairs were in good enough shape, so that saved us a little bit of money.
• Passed hor d’oeuvres on top of the cheese plates and veggies that came with our package. These were $2.50 per person.
• Cake/Dessert: You can easily get a sheet cake or some other dessert (cookies, donuts, etc.) for cheaper than what we paid for our cake, about $4 per person. Originally I had only budgeted $2 per person for this but in the end I wanted a larger cake. You can always buy a tiny wedding cake for the actual cutting if you want to do the ceremonious part without the cost. We did not buy a cake knife, we were supposed to use one supplied by the venue. Your venue might also have a cake cutting or plating fee, so look out for that. Luckily mine did not.

Attire

This section is pretty self explanatory. I severely underestimated the amount that I would spent on bridal accessories (another item I splurged on was my shoes, I spent $80). You can save money if you don’t have a veil, or wear cheap shoes, or buy a lightly used dress but I did not do those things.

We used Men’s Wearhouse and David’s Bridal for all our attire, which was really helpful since we had a lot of out of town people in our bridal party that needed to be able to get the same dresses/suits. My experience with these vendors was great and easy. The bridesmaid dresses might take a while to come in, especially if you want them ordered in not-as-popular colors, so it’s a good idea to order them early. I recommend at least 5 months before the wedding.

If all of your groomsmen get their suits from Men’s Wearhouse the groom will get his for free. Men’s Wearhouse supplies everything except your boxers (socks, shoes, belt, cufflinks, tie, etc) so that’s pretty nice for the guys. You can also include your fathers in the package at Men’s Warehouse if you want, which I found helpful.

Ceremony

We had a Catholic wedding ceremony and these prices were pretty standard for churches in my area. There is a lot of preparation that goes into getting married in the Catholic church (Pre-Cana), but all of that is free. The location fee was $315 and there was also another mandatory fee of $315 for the musicians that is down below in “Music”. We got married in a beautiful cathedral, so we didn’t need any ceremony decorations or accessories. We also chose not to include a unity candle in our ceremony, and the church already had reserved signs for the first couple pews that we could use for family. We had a family friend who is a priest and he performed our ceremony, so I just invented an officiant cost that I thought was appropriate and gave it to him in a card at the rehearsal dinner. I also gave him a bottle of whiskey as a gift.

Flowers & Decor

Flowers can also be a significant portion of your budget if you are not careful. Obviously you can save money by not having flowers, or using fake flowers, but I liked the look of real flowers enough that I wanted to do something simple with them. Another way you can save money is to make the bouquets or centerpieces yourself, but I did not want to do that because I was worried the flowers would wilt or die. Simple pink roses and baby’s breath made up all of our flowers except my bouquet. I tried to choose types of flowers that would be low cost, but in the end it’s just kind of hard to keep everything as cheap as I wanted and still get something decent. I used a local florist and was really happy with the results. I decided not to get any additional boutonnieres or corsages for parents or grandparents to save money.

I forgot to budget for a flower delivery fee, so that’s why my estimated costs are $0. If you have someone who can pick up the flowers you can avoid this, but probably you just want to pay it for the peace of mind.

I splurged on a really fancy cake topper with our last name on it, so I got a really simple guest book and card box (you can just get super cheap ones on Amazon or from Walmart and they still look nice).

Music

I budgeted $700 for a DJ and I have no idea how I picked this number. I ended up paying $750, and this was the cheapest DJ I could find. While their website is less than stellar, they actually ended up working out really well. The DJ was super nice and everyone had an amazing time. I also had a friend recommend this DJ beforehand so I had more confidence hiring him. We did not have a photobooth, or any kind of special lighting that some DJs provide.

There was a $315 ceremony musician fee which was required by my venue, a Catholic church. This paid for the cantor and the organist.

Photography:

I budgeted $700 for a photographer and $700 for a videographer. Again, no idea how I got these numbers. It became apparent pretty quickly that I was not going to be able to find a professional photographer to shoot my wedding for $700. If you know a friend with a camera that you want to trust/burden with the job of shooting your wedding, you could save some money here, but that was not the case for me. I decided to roll my videography budget into my photography budget and not have a video of my wedding. This was the right decision for me, but I get that it might not be the right decision for everybody.

Wedding photography can be EXPENSIVE. I saw packages upwards of $4000 for weddings with two photographers. If you have an extremely large wedding party or some huge distance to travel, I could see why you might want two photographers, but for the most part you should only need one. I was able to write up a schedule for the day that had the photographer everywhere they needed to be, and I did not feel like I missed any picture opportunities.

I found my photographer off of a recommendation from my sister-in-law. What I specifically loved about his package was that he gave us a digital copy of every single photo he took that day. He edited some important ones of family and the bridal party, and then gave us digital copies of ~600 other photos for us to do whatever we wanted with. There was only 1 flat fee, $1100 for 1 photographer the whole day. Once we had the digitals, we could spend any amount of money we wanted buying prints or books of the wedding. (In the end we bought 3 books from Shutterfly for about $180 total)

We did not have engagement photos. Some photographers will include this in with a wedding package, or charge $200-$500 for a session. We just took a picture with an iPhone for our Save The Date and it worked out great.

Stationary:

We got all of our stationary from Vistaprint, with the exception of programs and place cards. I think people mostly think of Vistaprint as a business printing store, but I found a theme on their site that I really fell in love with and it made for some super cute invitations. Unless you are really artsy, I would highly NOT recommend trying to make your own invitations. There are some kits that you can buy from Walmart or Target to print your own, but in my opinion that is more headache than it is worth.

Our place cards we got from Walmart, and I just wrote each person’s name on them myself. Our programs were pretty detailed and important because we were having a Catholic wedding ceremony, which are about an hour long. To create the program I used a combination of this link from the Knot and the planning sheet the church gave me to create what I wanted our programs to say. Then I tried to have them printed at Staples and fold them myself — big mistake! Maybe you can do this if you only have a one page program (but if you only have a one page program, why have a program at all?) With a multiple page program it is really hard to center it all and fold it correctly. I had Staples reprint my program three times before I finally gave up and ordered them online.

Finally, if you are sending save-the-dates, invitations, reply cards, and thank you notes, you need one stamp for each of those, which is four stamps at 50 cents a piece. To get the estimated cost for the rest of this section I am multiplying the cost of each of these paper items by the number of households invited to the wedding (80 in my case). 80 save the dates, 80 invitations, and 80 thank you notes. I actually probably ended up ordering 100 of each of these things because I was overboard nervous about writing the wrong address or messing one up.

I bought a stamp for our return address (and I also used it for the real address of the reply cards) which made it a little faster to put the invitations together and address the thank you notes.

Wedding Rings:

Our wedding bands were plain white gold bands that we got online. Typically a woman’s wedding band will be cheaper than a man’s because it is smaller. Here is an example of something similar to what I got. I thought about getting a band with diamonds or sapphires in it, but in the end opted not to so that I would have something simple to wear everyday.

Transportation:

Originally we were not going to have a limo for the wedding party but my family thought it would be really fun and helped pitch in above what they had already volunteered to contribute to the wedding. This was super fun, but not necessary. I think we used the same company that I used for prom, and it was a pretty standard package. We did not have any guest transportation even though our ceremony, reception, and hotel were in three different places, and at the time Uber and Lyft were illegal in upstate NY so our guests were pretty much on their own. We helped coordinate for a few people to get rides from others but for the most part people just took care of themselves. We also had family volunteer to pay for our overnight hotel room, but sometimes if you get married in a hotel they will include this as part of the package. The night before the wedding we each stayed in our individual parent’s houses for old time’s sake (rather than two individual hotel rooms).

Gifts:

Lastly, gifts. I gave gifts to my bridesmaids when I asked them to be in the wedding, and we gave gifts to our entire bridal party at the rehearsal dinner. We also gave cards and nice notes to anyone else who was helping us with the day. We gave gifts to our parents which we ended up spending quite a bit on. For favors we bought tiny jars and filled them with pink candy to match our color scheme. We wanted to use actual M&Ms because M is the first letter of our last name, but using “the real thing” is pretty expensive. We ended up being able to get on-sale pink M&Ms the week after Valentine’s day which saved us a ton of money! But you could always do this with cheap candies as well. We also got a hotel room block at a local hotel (the Marriott). This gave a discounted rate to any of my guests that were coming in from out of town, but didn’t actually cost me anything. I created hotel gift bags for the hotel to hand out to guests as they checked in. They mostly included snacks from our favorite local grocery store, Wegmans! They also included some “hangover” medication and Do Not Disturb door signs that said “Shh… we are recovering from Shannon and Tristan’s wedding!” These gift bags were definitely a splurge item and not an essential, but they were so cute and fun that I’m glad I spent the extra money on them.

The Total

Drumroll please! At the end of the day, we went about $1500 over budget. The total cost of our wedding was $16,462.96. Based on my expenditures above, I am very confident that we consciously chose to go over budget. We knew when we were going over budget. We made these decisions based on things we knew we really wanted, and also extra money that we got from engagement party presents or from family.

The theme of our wedding was #BestDayEver and we absolutely had the best day ever! I still hear from family and friends that they had so much fun at our wedding and if I had it all to do again, I wouldn’t change a thing (except for maybe not trying to print the programs myself…lol). According to The Knot, Americans spend an average of $35,329 on their wedding which just seems like insanity to me. The average cost of a wedding in my area of New York is estimated at $27,302. We did not go into any debt for our wedding, and we were able to go to Hawaii for our honeymoon for $4000. Good luck on your own wedding planning journey!