<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by octal on Medium]]></title>
        <description><![CDATA[Stories by octal on Medium]]></description>
        <link>https://medium.com/@octal?source=rss-6e5a733b5107------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*S-4CjY_pO3ILXOov.jpg</url>
            <title>Stories by octal on Medium</title>
            <link>https://medium.com/@octal?source=rss-6e5a733b5107------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 01:44:52 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@octal/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Cloudbleed: How to deal with it]]></title>
            <link>https://medium.com/@octal/cloudbleed-how-to-deal-with-it-150e907fd165?source=rss-6e5a733b5107------2</link>
            <guid isPermaLink="false">https://medium.com/p/150e907fd165</guid>
            <category><![CDATA[security]]></category>
            <category><![CDATA[cloudflare]]></category>
            <category><![CDATA[taviso]]></category>
            <category><![CDATA[passwords]]></category>
            <dc:creator><![CDATA[octal]]></dc:creator>
            <pubDate>Fri, 24 Feb 2017 01:47:57 GMT</pubDate>
            <atom:updated>2017-03-15T15:32:40.944Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/712/1*i4Lt-6kqZLdqGFXyLHr5Vg.png" /></figure><p>Tavis Ormandy (<a href="https://www.twitter.com/taviso">Tavis Ormandy</a>) of Google’s Project Zero uncovered a major vulnerability in the Cloudflare Internet infrastructure service. Essentially, web requests to Cloudflare-backed sites received answers which included random information from other Cloudflare-backed sites! This information could potentially include confidential information (private messages on dating sites, emails), user identity information (Personally Identifying Information (PII), and potentially in a healthcare context, Protected Health Information (PHI), or user, application, or device credentials (passwords, API keys, authentication tokens, etc.)</p><p>Both Project Zero and Cloudflare acted promptly. The bug was reported on 2017–02–17 and a mitigation was in place within an hour. Public notification was given on 2017–02–23.</p><ul><li><a href="https://bugs.chromium.org/p/project-zero/issues/detail?id=1139">1139 - cloudflare: Cloudflare Reverse Proxies are Dumping Uninitialized Memory - project-zero - Monorail</a></li><li><a href="https://blog.cloudflare.com/incident-report-on-memory-leak-caused-by-cloudflare-parser-bug/">Incident report on memory leak caused by Cloudflare parser bug</a></li><li><a href="https://news.ycombinator.com/item?id=13718752">Ormandy: Cloudflare Reverse Proxies Are Dumping Uninitialized Memory | Hacker News</a></li></ul><p>The duration (2016–09–22 to 2017–02–20) and potential breadth of information exposed is huge — Cloudflare has over 2 million websites on its network, and data from any of these is potentially exposed. Cloudflare has said the actual impact is relatively minor, so I believe only limited amounts of information were actually disseminated. Essentially, broad range of data was potentially at risk, but the risk to any individual piece of data was very low. Regardless, unless it can be shown conclusively that your data was NOT compromised, it would be prudent to consider the possibility it has been compromised.</p><p>While Cloudflare’s service was rapidly patched to eliminate this bug, data was leaking constantly before this point — for months. Some of this data was cached publicly in search engines such as Google, and is being removed. Other data might exist in other caches and services throughout the Internet, and obviously it is impossible to coordinate deletion across all of these locations. There is always the potential someone malicious discovered this vulnerability independently and before Tavis, and may have been actively exploiting it, but there is no evidence to support this theory. Unfortunately, it is also difficult to conclusively disprove.</p><p>The most sensitive information leaked is authentication information and credentials. A compromise of this data can have lasting and ongoing consequences until credentials are revoked and replaced.</p><p>From an individual perspective, this is straightforward —the most effective mitigation is to change your passwords. While this is on all probability not necessary (it is unlikely your passwords were exposed in this incident), it will absolutely improve your security from both this potential compromise and many other, far more likely security issues. Cloudflare is behind many of the largest consumer web services (Uber, Fitbit, OKCupid, …), so rather than trying to identify which services are on Cloudflare, the most cautious is use this as an opportunity to rotate ALL passwords on all of your sites. This will improve your security, although the primary benefit is from threats unrelated to this incident.</p><p>(Best practice is to use a long random string for each password, unique for each site, and to manage that collection using a “password manager”, such as 1Password, LastPass, or the built-in password managers in modern web browsers. Users should also log out and log in to their mobile applications after this update. While you’re at it, if it’s possible to use 2FA or 2SV with sites you consider important (using something like TOTP/Google Authenticator or U2F), that’s a meaningful security upgrade, too.)</p><p>For site operators using Cloudflare: while it may be disruptive, you should seriously consider the impact on your users. Use this incident as an opportunity to exercise your incident-handling process — discuss the specific impact to your application and what response makes the most sense (it is likely different for every site or application.) You have to balance an unknown but small risk vs. other costs. Be careful not to “scare” them, but it may be prudent to take action. At minimum, I would prepare a “stock answer” for support about this incident, and in an enterprise or b2b context, proactively communicate with customers about the extent of the incident and its effect on your application and their data. For the vast majority of sites on Cloudflare, that alone is likely sufficient.</p><p>There are very real costs to forcing a user password change — users might lose trust in your service, and it’s an inconvenience. It doesn’t appear large numbers of credentials have been compromised, so for a consumer service with limited risk to compromised accounts, it may not be worth the effort to mass-invalidate passwords and force a change. For administrator credentials, or for any sites processing highly sensitive information through Cloudflare, the lack of a quantifiable maximum exposure probably means it is worth forcing a password update. If you had any other reasons for wanting to push a password update to all of your users, this would of course be an additional factor in the decision.</p><p>Sites should invalidate authentication credentials for mobile applications and other machine to machine communications (IoT devices, etc.), forcing users to re-enroll apps and devices, if they used Cloudflare as an infrastructure provider. If you don’t want to force a change of passwords, a viable intermediate step could be to invalidate authentication tokens, forcing users to log in again with their existing passwords; since authentication tokens are more frequently passed between browser and server, they are far more likely to exist in a leak of random memory from the server than raw passwords, and forcing a fresh login is a minimal impact to users and may not require any specific messaging or notification.</p><p>Additionally, any sites not on Cloudflare but with a lot of users who use Cloudflare-hosted sites (essentially any large or consumer sites on the Internet) should consider forcing user password changes in case their users had re-used the same passwords on each site. This is probably not warranted for the vast majority of sites (anyone with security needs so high that this would make sense should be using authentication infrastructure far stronger than user passwords; otherwise, it is nearly certain users will re-use passwords with a compromised site), but it may be a possibility. Personally, I would use it as an opportunity to upgrade my overall authentication infrastructure in such a case — deploying 2FA (ideally U2F or better, or user-selectable TOTP/HOTP, not SMS), account activity monitoring, etc. — rather than rushing a change like invalidating passwords. Standalone passwords on the Internet are no longer best practice for user authentication.</p><p>If your application or website is on Cloudflare and is subject to industry or national regulation, this may be a reportable incident. (examples would be health care/medical privacy concerns with HIPAA). Your security and compliance teams should evaluate. Obviously, full compliance with applicable regulations is an essential part of security.</p><p>Subjectively, I believe this is a serious security issue and should be evaluated by any major service using Cloudflare. It is likely not an “end of the world” in severity, as unless there was concerted malicious exploitation, vulnerable data is probably fairly randomly distributed across the Internet, but presence of data in searchable caches might make small-scale exploitation possible. Since the mitigation for end users is good advice in general (use a password manager, use unique random passwords per sites, rotate on any compromise), it’s a “no brainer” for most security-conscious end users. For site operators, the decision really comes down to your specific userbase and their level of security sophistication and risk tolerance. I’m deeply appreciative of both the ongoing work by Google’s Project Zero (especially Tavis) and by Cloudflare’s rapid response to this issue.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=150e907fd165" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Review of In Context: Understanding Police Killings of Unarmed Civilians]]></title>
            <link>https://medium.com/@octal/a-review-of-in-context-understanding-police-killings-of-unarmed-civilians-165d1c496fff?source=rss-6e5a733b5107------2</link>
            <guid isPermaLink="false">https://medium.com/p/165d1c496fff</guid>
            <category><![CDATA[public-policy]]></category>
            <category><![CDATA[police]]></category>
            <category><![CDATA[police-shooting]]></category>
            <dc:creator><![CDATA[octal]]></dc:creator>
            <pubDate>Sat, 30 Apr 2016 03:49:44 GMT</pubDate>
            <atom:updated>2016-04-30T03:49:44.661Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="http://www.amazon.com/Context-Understanding-Killings-Unarmed-Civilians-ebook/dp/B01DO9NTAG/">A</a> Review of <strong>In Context: Understanding Police Killings of Unarmed Civilians</strong></p><p><a href="http://www.amazon.com/Context-Understanding-Killings-Unarmed-Civilians-ebook/dp/B01DO9NTAG/">This is a review of In Context: Understanding Police Killings of Unarmed Civilians</a>, a book by <a href="https://twitter.com/nselby">Nick Selby</a>, Ben Singleton, and <a href="https://twitter.com/EdFlosi">Ed Flosi</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/334/1*mVqFkgItypmGCTaSmgTDlQ.png" /></figure><p>In Context is an excellent case-by-case analysis of each US police killing of unarmed civilians in 2015. Three law enforcement experts (the authors) used public sources, their own training and experience, and a solid understanding of the law to give opinions on each of these unfortunate incidents. (Be prepared to be upset in reading this book, especially the cases; every one of them is unquestionably sad.)</p><p>Police killings of unarmed civilians (a specifically defined group — the presumption is that these incidents could be the most egregious uses of force) are documented more than most police/civilian interactions in the press, and have in many cases been major news stories both locally and nationally. This is a major political issue now — some people believing police are killing out of murderous or racist intent, other people denying that every civilian death is completely unavoidable and entirely the fault of the deceased. The only thing these sides have in common is that neither argues from data, and both have pre-existing ideas about what the causes and solutions are. Reality is far more complicated.</p><p>By doing the hard work to assemble the data and analyze all of these cases, the authors have uncovered the reality of the problem — police killings of unarmed civilians are not a monolithic set, but several distinct types of cases. There is no single solution, but several major categories can be addressed in turn. The raw data set is available as the <a href="http://www.streetcredsoftware.com/pkic/">StreetCred Police Killings in Context dataset</a>.</p><figure><a href="http://www.streetcredsoftware.com/pkic/"><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xKoutdS0pmnpq_mlSjPNlg.png" /></a></figure><p>The authors are all experts — I have a reasonable degree of familiarity with the law of self defense from a civilian perspective, having taken about 200 hours of range and classroom training from Front Sight, Massad Ayoob, and others, as well as experience in combat zones in Iraq and Afghanistan as a contractor, and many friendships with lawyers, law enforcement, and self defense experts — and I was pleased to see that my conclusions were fairly similar to theirs. I know Nick Selby from technology law, so I already understood his position, but after reading about 10 cases, realized I agreed with the other two authors to a similar degree. They are not, as one might expect from police officers, blindly supportive of police — they are certainly respectful of officers and their job, but did not hesitate to criticize and critique incidents where things could be improved. All three of them are exactly the kind of people I’d like to see responding to an incident if I call for service.</p><p>The first point from reading the book (and forming my own opinion on the data, although largely the same as the authors) — police killings of unarmed civilians are thankfully a relatively low number. Obviously the only acceptable number is zero, but out of the millions of police interactions per year, often at the worst moments of someone’s life, 153 is lower than I would have expected. However, these incidents, while they receive a higher amount of press coverage than most incidents, do not have any central reporting or cataloging of data, so there are incidents in which public data is insufficient to make an accurate evaluation. Given the critical severity of these incidents, and the societal benefit to having the citizenry oversee the actions of its paid police forces, I believe there should be a formal reporting and investigation procedure, independent of the involved agency, for every death involving the police, particularly of unarmed civilians. Unfortunately, local news media are not necessarily in a position to uncover and publish the truth of these incidents, and there is a substantial interest in aggregating all of these incidents and generalizing trends, rather than focusing only on each incident in isolation.</p><p>A second big takeaway from this book is that in a lot of these cases, it was a “police attended death”, more so than something intentionally caused by the police. In many cases, a person has taken drugs/alcohol to excess, or has an underlying mental or physical disorder, or is otherwise in a state of trauma, and then dies subsequent to police interaction. In some of these cases, police used non-lethal force (TASER, physical restraint) and the civilian dies, but it is very unlikely the force used caused the death, and even less likely that the force used was reasonably believed to be likely to cause a death. There has to be a solution to this — perhaps better emergency medical training for police, perhaps technology — but it’s not a case of police intentionally doing things to cause death. My biggest criticism of policing is that police are not equipped and required, when force protection/officer safety would allow, to provide the same level of immediate lifesaving aid they would provide another officer to any civilian, including one with whom they just had a use-of-force encounter with.</p><p>A subset of this is that mental health and substance abuse are critical problems in America, and must be addressed far upstream of violent interactions with the police. This is a much bigger problem than just police killings of this group — even absent that, their lives are often nasty, brutish, and short, and as a country, we owe them something far better.</p><p>Third: police should be more willing to use a wider variety of force, rather than progressing immediately to deadly force once one non-lethal force option has failed. The most egregious example seems to be the TASER — in several incidents, officers seemed to only be choosing between an TASER and a firearm, not considering the use of physical force, baton strikes, OC spray, or other implements. There are many potential causes for this, but they all come down to training, standards, and leadership. If a force is not trained to behave in a certain way, and if officers are not held accountable to that standard, they won’t behave that way when needed. While it’s viscerally revolting to see a police officer use physical force to subdue a suspect, and while this may lead to injuries to both suspect and officer, it’s far preferable to an officer progressing to deadly force when deadly force could otherwise be avoided. There are obviously cases where progressing immediately to deadly force, or going directly from one non-lethal to one deadly-force options is the correct response, but there appear to be some avoidable civilian killings where alternative non-lethal force may have been an option.</p><p>Fourth: it appears police are often putting themselves at more risk than they must, and this leads to avoidable civilian killings as well. In cases where there’s no immediate risk to life of the officer or other civilians, it is far better to contain a threat and bring additional resources to bear (overwhelming force, technology, or simple exhaustion on the part of the civilian), rather than to engage. There was one especially horrible incident where thieves had broken into an empty structure at night, triggering an alarm, and the officer entered to clear the alarm, rather than waiting for backup, which turned into a struggle and then shooting. That could have been avoided if he had cordoned the area and waited for backup.</p><p>Fifth, there is an opportunity to use better technology to both reduce the number of civilian killings and to protect police and make them more effective. While there has been criticism of “militarization of police”, continued research and development of non-lethal force options, improved data gathering tools (police-worn body cameras, most acutely), and possibly better tools for medical and psychiatric intervention during and after incidents (such as simulators in training, and improved monitoring and first aid tools deployable by police) could save lives.</p><p>From what I saw of these incidents, I didn’t see any particular racial or other bias on the part of the officers, at least in the overwhelming majority of cases, leading to police killings. However, minorities are unquestionably on the receiving end of many of these incidents. Why?</p><p>For those who believe “the system” is racist — I agree. However, at the point of police use of deadly force, the system is the most mechanical and most scrutinized, and thus the least likely to show racial bias, so the correct place to look is upstream of these incidents. Drugs, poverty, and mental illness are clearly the overriding factor in many of these incidents, and those problems disproportionately affect minority individuals and communities. Addressing those problems, and any unique factors (employment discrimination, geography, lack of access to education and health care, etc.) would lead to fewer police killings of unarmed minority civilians (and to fewer killings overall).</p><p>Police use of force in incidents which do not result in killings, and use of force in situations other than by police (such as by corrections officers in prisons) were not addressed in this book, and themselves are also far more likely to have a racial or discriminatory component. Police engaged in the use of deadly force are certainly aware that they will be “under a microscope” after the incident, and are acting according to training. Authorities in other contexts, with more time to act, less certainty of observation, and overall lower stakes for themselves are far more likely to engage in out of policy violence.</p><p>In short — racism by security forces is more likely to occur when a minority suspect is unnecessarily stopped, more forcefully searched, or otherwise not given the same status as any other citizen, than when a police officer is at the point of using deadly force. And it is far more likely still to occur when a minority family must live in a high-crime neighborhood due to poverty and lack of access to jobs and education.</p><p>Overall, In Context is an excellent review of the available information on every police killing in the US of an unarmed civilian in a single year, and some excellent policy ideas on how to improve policing and civilian oversight of policing in the future. Hopefully it will lead to additional, data-based analyses of other forms of police interactions, and to substantive improvements in society.</p><p><a href="http://www.amazon.com/Context-Understanding-Killings-Unarmed-Civilians-ebook/dp/B01DO9NTAG/">In Context is available from Amazon.com as a Kindle e-book.</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=165d1c496fff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Maximizing your chances of getting into Y Combinator (Summer 2014)]]></title>
            <link>https://medium.com/@octal/maximizing-your-chances-of-getting-into-y-combinator-summer-2014-b0a75aa7401f?source=rss-6e5a733b5107------2</link>
            <guid isPermaLink="false">https://medium.com/p/b0a75aa7401f</guid>
            <dc:creator><![CDATA[octal]]></dc:creator>
            <pubDate>Wed, 19 Mar 2014 08:27:50 GMT</pubDate>
            <atom:updated>2014-03-19T20:04:17.215Z</atom:updated>
            <content:encoded><![CDATA[<h4>Part 1: The Application</h4><p>Y Combinator Summer 2014 applications are due 28 March 2014 at <a href="http://ycombinator.com/apply.html">http://ycombinator.com/apply.html</a></p><p>The written application with video is the first step in a multi-stage process. The ‘zeroth’ stage is deciding to apply at all. Presumably if you’re reading this you’re already interested in Y Combinator; if you haven’t made up your mind, there are various articles on the Internet about this topic. As a founder from YC S11, I strongly believe Y Combinator adds a lot of value to most Internet startups, but is clearly not essential to founding a successful startup. Even if you don’t ultimately do YC, the application process tends to be very beneficial for clarifying how you think about your startup.</p><p>Make sure you apply at the right time for your company. In a lot of cases, the greatest benefit from Y Combinator comes when you can launch your product at least a month before Demo Day (at the end of August, for Summer 2014), so you can show traction to investors. Even if I’d decided to go through Y Combinator, if I didn’t think I would be ready in time, I might consider waiting until the following batch to apply (late October 2014 for Jan-March 2015 Winter 2015).</p><p>Once you’ve decided to apply, the next step is to submit the written application with video. It helps to understand how this application gets used. First, it’s used for quick-pass screening of applicants by the YC partnership. Next, it’s used for more in-depth scrutiny of which groups will be given interviews. Finally, it’s used for preparation by the YC partners before interviewing selected groups (both to determine which partners interview which teams, and to figure out which questions they might want to ask). The same application needs to be a work for all three steps.</p><p>If you submit a substantially blank or incomplete application (“wow, this is horrible”), it’s probably not going to receive serious consideration at this stage — there are just too many applications to go through. Obviously, don’t click submit (vs. update) until your application is complete and ready for review! If an application is written so badly as to be difficult to understand when quickly scanned by a reader, it also has serious chances of being rejected. If you’re not a fluent writer in English (which is possible even if English is your native language, sadly), you’ll benefit from a review by someone who isn’t intimately familiar with your product, just to make sure it’s clear. Obviously, you want to avoid falling into this category.</p><p>After the “horrible” applications are discarded, exceptional applications which immediately stand out get fast-tracked to the next stage. If you have a team of people who have accomplished huge things in the past, with a product already launched, an excellent demo, and a clear video, there isn’t much to worry about. If you’ve worked for a Y Combinator company, know the YC partners, and are doing something amazing, you probably don’t need too much help with your application. Good luck, and get ready for the in-person interview stage which follows. You probably know if you fall into this category already. The best advice for this category is to make sure factors which make you exceptional show up quickly during a cursory scan, just to make things easier for the people reviewing a huge stack of applications.</p><p>If an application on its face seems incredibly bad (blank, misleading, offensive, etc) but has an amazing live site with lots of users hidden behind a link, it’s hard to fault people for not clicking through. Your responsibility as an applicant is to put your best application in front of the reviewers; your ability to do that is part of the test. It’s useful to note how much time might be spent reviewing an application in this early stage. In general, it’s a very bad idea to make your application depend on someone installing an app on a specific mobile phone OS (or really any installation at this stage), or signing up for an account, or even depending on someone watching your video or viewing images. (Even for purposes other than the YC application, I’d strongly suggest a good textual summary of content, followed by static images, followed by video or a login-less web link, followed by a live website, followed by an installable free app, followed by a paid mobile app, if applicable.)</p><p>Now we are left with the middle category. This category is the majority of applications. These consume the majority of time when reviewing, as the reviewing process is biased to avoid false negatives (rejecting a team early on who might be good), but is also very high volume. Ideally, you won’t be in this category, but realistically, a lot of applications will be, so this is where editing and tips provide the greatest benefit.</p><p>The video (with some YC recommendations and good sample videos at <a href="http://ycombinator.com/video.html">http://ycombinator.com/video.html</a>) is probably most important in the early screening stage for these “middle” applications — it’s entirely possible it never gets watched for the really bad ones, and possibly for the best applications (because the decision can be made so quickly). There are some key elements to make the video as good as it can be:</p><ol><li>Technical quality of the audio is much more important than technical quality of video. Avoid wind noise, avoid fans and low-quality built-in computer microphones, and listen to your recording a few times to make sure it’s clear. Cellular phone wired headsets often have particularly good microphones relative to computers or other devices, or dedicated audio recording setups like the $50 Blue Snowball USB mic. On the other hand, a low quality computer webcam is probably fine for the video, although positioning it so you aren’t looking down at a laptop on your lap, with reasonable room lighting, etc. can be helpful. Absolutely do not pay for professional video services, though — too slick a presentation usually turns people off as well, and it’s a waste of money.</li><li>If you’re not a native US English speaker, be careful of your accent. Paul Graham got pilloried over faulting heavy and incomprehensible accents, but especially in a rapid screening environment, it makes a big difference in how intelligible you are. Speak more slowly than you would otherwise if your accent is likely to be unclear.</li><li>Submitting a video you made for another purpose (Techstars application, some local news story, a trade show, a canned sales pitch) is pretty likely to get your application rejected out of hand. Spending 15 minutes on a quality video specific to the needs of the application is a reasonable requirement for a successful application.</li><li>Don’t read from a script. Do take a few takes if necessary to get the video to be smooth and cover what you want to cover, and make sure you review it.</li><li>Smile. Most of the value of the video is showing who you are, and that you are comfortable talking about what you’re doing in front of a (virtual) audience. Ideally, also show your team dynamic, but having one person doing the majority of the talking can be fine (although it shouldn’t be entirely one person).</li><li>Ideally, the entire team should be in one place. The “Skype” or “Google Hangouts” videos are always weaker than videos where the team is in one place. If it’s unavoidable, perhaps you should have applied sooner (but go ahead anyway, just know it might hurt your chances, and try to make sure the rest of your application is as strong as it can be.)</li><li>Don’t go on for too long, and put your most important message up front (and repeat it at the end). If your biggest strength is that you have a 100 site business deployment underway, lead with that in the first 15-30 seconds, rather than a long discussion of your technology or the problem — you can circle back through the problem and your technical solution and team, and then emphasize your deployment strength again at the end. You should have your message communicated within 15-30 seconds, and then spend up to another minute elaborating on that and the rest of your message; under no circumstances should your video be over 2 minutes, and ideally it would be about 1 minute.</li><li>Set it on YouTube to be “embeddable”, but not “listed”, and not “private”. The YC application review process embeds the videos within a reviewing page, so it’s inconvenient for reviewers when they’re submitted on vimeo or other sites, or when embedding is disabled.</li></ol><p>Once you have the video, there are a lot of little details which might make a difference if your application is on the line.</p><ol><li>Submitting functional, correctly-formed (and thus clickable) URLs where requested</li><li>Correctly following directions for listing YC usernames (i.e. make sure you include your own username and that of your co-applicants)</li><li>Following the (slightly complex) instructions for founder credentials. If someone isn’t smart enough to follow basic formatting instructions, I often question that person’s general intelligence. Use semicolons as instructed. Use standard abbreviations. Include city or country name in parentheses if you’re referring to an institution like “National Technical University”. Separate founders with blank lines.</li><li>Correctly answer the “yes or no” dropdown about whether you’re a single family, the team can’t move to the Bay Area, etc.</li><li>If a question isn’t applicable (e.g. “how do you know your cofounders” as a single founder team, don’t get cute — just write “single founder” or “n/a”.</li></ol><p>Now you’re on to the real content of an application — your answers to the various short-essay questions. Different reviewers have their own opinions on the most important questions, but the key to the whole thing is that none of it is a “trick” — the goal is to communicate why you’re a great team as clearly as possible every time the application is reviewed at each step in the process.</p><p>One way to accomplish this is to “write like a newspaper” — don’t “bury the lede”. i.e. you should put the most important facts in the first sentence, and then explain them. If the question is “How long have the founders known one another and how did you meet? Have any of the founders not met in person?”, don’t go into a paragraph long essay/wall-of-text about how everyone is connected through the Internet these days but a couple of people have similar interests in sciences and went to a really good local lab space and then happened to meet up in the meeting about kicking out the bio guys who were stinking up the lab but you also have an interest in biology from back in college. Just say “We’ve met in July 2012 at a local hackerspace while working on personal projects, and collaborated on two small open source projects (URL and URL) before this one.” Think about why the question is being asked, and try to answer the underlying question. In the case of “how did the founders meet” (one of the easiest questions), it’s to see how deep your connection is (i.e. you didn’t meet up a couple days ago just to work on the YC application, and you’re not just roping someone in due to YC’s bias against solo founders…). Make sure your answer addresses the underlying concern.</p><p>Personally, I find the “what is your company going to make”, “most impressive accomplishments”, “how did you pick this idea”, and “what are the alternatives” and “competitors” questions most meaningful and predictive. Put a lot of effort into these — saying as clearly as possible, in as few words as possible, with as little reader preknowledge demanded, exactly what you want to say. Do your homework; quantified data about size of markets, specific alternatives and competitors, technologies, etc. make you seem knowledgeable and convincing, and also help you make your case (both are important).</p><p>Showing that you understand research is important and are willing/able to spend a few hours researching the topic you’re allegedly willing to devote the next 10 years of your life to disrupting is probably a necessary part of a successful application.</p><p>It is very helpful to have multiple people review what you write for these questions. One common mistake is assuming the reader is already an expert in your field and familiar with the trends and technologies within it (i.e. for a radiology PACS-as-a-Service, you probably can’t assume the reader knows what clinical radiology departments are like, or what a PACS is or why they’re useful); another mistake is presenting an argument without sufficient reference to industry-standard terms, so a reader who is familiar with the field might think you are not familiar with the industry.</p><p>Being able to clearly and succinctly convey your message is a valuable skill in entrepreneurship in general, not just on the YC application. If it takes a couple days to get the answers to 5-10 questions down to a few exceptionally clear sentences each, it’s time well spent.</p><p>A lot of the questions exist primarily to disqualify you; the default is that they’re not harmful. Equity distribution within a certain range isn’t a big deal, but a 90/10 business/hacker split generally means the businessperson is a douchebag and the hacker is uninformed — one of those is a lot easier to address with 5 minutes on the Internet than the other. The general assumption is “you’re doing this project anyway, and would like to get into YC if possible”; a project you will do if and only if you get into YC is probably not going to get into YC — so the “what level of commitment for the next year”, “other commitments”, etc. questions should be fairly clear (there are exceptions if you have a non-primary founder with some side obligation, especially in a three or more person team). If you don’t own or control the IP necessary for your business, you’re also possibly doomed, but this is rarely an issue. Participation in most other incubators or accelerators is generally a black mark; the exception is college-specific incubators like START while a student. (I don’t know of any Techstars alumni who have gotten into YC with the same company.) Generally, the incorporation, “where you will be based after YC”, etc. stuff won’t hurt you.</p><p>“Other ideas”, “hacked a system”, and “tell us something surprising” are kind of wildcards — if you have a great answer, they can help. Leaving them blank or entering something moronic will hurt, but a basic answer should not affect your application.</p><p>There’s a whole class of questions which are incredibly important but essentially obvious: the quality of any demo, and “how far along are you”. It’s pretty obvious — putting effort into a demo (or working app), and being as far as long as possible, is going to make your application stronger. You don’t need a blog post to tell you that.</p><p>I’d probably recommend filling in your application and doing a video, then saving it and sharing it with anyone willing to review your application. You will get a lot more meaningful feedback than if you just say “hi, I want to apply to YC, do you have any advice” — both because people are more likely to blow you off if you’ve not made any serious effort yet, and because it’s difficult to give anything but the most general advice without seeing your answers to the questions themselves. If you’re asking people you don’t know well, the best way is a 1-2 sentence introduction to yourself and your product, a clear ask (ideally just around a specific question in the app and your answer, vs. the whole thing), and the text inline. Anyone worth getting advice from is probably quite busy, so keep that in mind.</p><p>Once you do click submit, don’t stop working! One common outcome is that an application gets past initial review, but before being selected for interviews, one of the partners contacts the team and asks some questions (perhaps about the demo or something else) — if you can show that you’ve made progress toward whatever issue they raise, even if it’s not completely solved, you’re in a much stronger position. Plus, if you do get an interview, you’ve got to go through that hurdle, and then ultimately build the company, so this is just one early step on a very long road.</p><p>Just to talk about numbers: from public figures, YC gets 2000 to 3000 applications per batch, and accepts a couple hundred groups for interviews, so maybe 10% pass the first stage. Once you’re accepted for an interview, around 50-70 groups are admitted to YC, so maybe 30% of those with accepted applications pass the second stage interview. Not getting an interview doesn’t hurt your chances of getting an interview when you re-apply, and being invited to interview (but not accepted) generally seems to be a positive factor for future admission, particularly if any issues raised during the interview are addressed.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b0a75aa7401f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>