HNHacker News
TopNewBestAskShowJobs

ptoomey3

311 karma · joined April 15, 2009

submissionscomments
ptoomey3··on Ask HN: How do you store photos and videos?
I just tried out the shared album thing. I shared a photo whose original size is 3.4MB. The photo you get when you download it from the shared album on iCloud is 925KB. And, I'm pretty sure there is no checkbox you can check to prevent that ‍️. I love lots about the apple ecosystem, but their photo management/sharing system is not one of them. It boggles my mind, as lots of Apple engineers have families. But, despite that, they don't have any system for seamlessly sharing photos (or entire libraries) with family members.
ptoomey3··on Ask HN: How do you store photos and videos?
Manually sharing works ok for super casual use, but it is way too much work when you take pictures daily. Plus, you would need to share photos both directions. And, even if you do all that, which photo library is the “source of truth”? Also, it is my understanding that photos shared through apple’s photo sharing system are not original photos or videos (they are shared at a lower resolution).
ptoomey3··on Ask HN: How do you store photos and videos?
If you want them, let me know and I can drop my hazel rules somewhere. They aren’t wildly complex, but it can be slightly confusing if you’ve never used hazel before. Once you see the rules it‘s super obvious what’s going on.
ptoomey3··on Ask HN: How do you store photos and videos?
Yeah..totally. I only meant “server” in so far as some machine where you want the “source of truth” to be. I personally want my source of truth to be my home iMac since it has a big drive that gets real-time backed up to backblaze. So, for me, I view that iMac as my “server”. In truth it can be any machine you are willing to leave an account logged into.
ptoomey3··on Ask HN: How do you store photos and videos?
Also, while talking up google photos, I must note the home hub is worth the price of admission even if all you use it for is as a photo frame for google photos. I’m mostly an all apple house (HomeKit, etc). But, google photos plus the home hub has been a huge hit with the family.
ptoomey3··on Ask HN: How do you store photos and videos?
While I voted for google photos in https://news.ycombinator.com/item?id=18770681, I will say that apple photos is nice if you have zero need to share a library with family members. For better or worse, apple Unlike third-party uploaders, Apple photos can “cheat” and upload no matter what. That by itself makes it the most reliable for Apple ecosystem folks. That said, I’ve not had any issues in recent history with the google photos uploader.
ptoomey3··on Ask HN: How do you store photos and videos?
This! I’ve tried every option under the sun at least twice and google photos wins. My current system has been this for the past 2 years or so:

1. Create a family gmail account (ex. house-yourlastname@gmail.com)

2. Install Google Photos on the both you and your spouse’s phone and set the automatic photo upload to upload to the house-yourlastname@gmail.com account.

3. Create a new account on your local home server (ex. house-yourlastname)

4. Install and login to the google drive client for the house@yourlastname.com account. DON’T do any of their photo upload nonsense. Just use the app to sync your google drive contents.

5. Go to https://photos.google.com and set the "Sync photos & videos from Google Drive” to “true”.

6. Go to https://drive.google.com and set the "Automatically put your Google Photos into a folder in My Drive” to true.

7. Install hazel on your home server and have it monitor the automatically created “Google Photos” folder that shows up in your Google drive (this folder shows up as a result of step 5). This folder is where all auto-uploaded photos from your phone(s) end up by default.

8. Have hazel move all auto-uploaded photos out of the “Google Photos” folder in Google Drive and into a folder structure of your choosing in Google Drive (Ex. “Google Drive/Photos/2018/11/2018-11-25-1234.jpg”). Because of step 4 the photos we move into our own folder structure will show up in https://photos.google.com.

9. Copy fancy camera photos into the same folder you are monitoring with hazel. The same hazel rules will rename and place photos into the same Google Drive directory structure.

10. Done - you have a pretty good shared family library. All family members can view all photos using the google photos app OR by visiting https://photos.google.com. And, any management (deletes, etc) are propagated to Google Drive. For example, I’ll triage on https://photos.google.com and all deleted photos end up n the trash can of my iMac by the time I happen to check.

Some general reasons why I landed on this:

1. I have a family and whatever I chose had to work for the whole family without jumping through manual copying at home. This basically nixes Apple photos, as they haven’t yet figured out we have families.

2. The only graceful way I’ve found to have a family library is to have spouses share an account. You could do that with an iCloud account, but there are lots of downsides to that. Google apps support multiple account gracefully, which allows you to be logged into a different account for Google Photos vs. Gmail (or whatever).

3. Most photo management solutions require you go sit down on the “shared family computer” to manage them. Google Photos bootstrapped off of using Google Drive lets you manage photos from anywhere (your phone, your spouse’s phone, the website, etc). The changes are synced back to Google Drive and back to your “source of truth” on your home server, Synology, etc.

4. The above solution only relies on the home server for sake of moving photos around to the folder structure of your liking. Even if your machine is turned off for a week, all your photos end up in the family library pretty quickly. Whenever you happen to start the server again, all the photos sitting in the “Google Photos” folder will then get organized into your preferred folder structure. This is appealing, as I didn’t like any solution I tried that relied on a server being up 24/7 to have all photos show up in the family library.

ptoomey3··on Zip Slip Vulnerability
Ha..indeed. We wrote something up about it in 2009: https://labs.neohapsis.com/2009/04/21/directory-traversal-in.... But even in that post we note that it was a relatively well known issue.
ptoomey3··on Cure53: Browser Security Whitepaper (2017) [pdf]
And just in case folks aren't super familiar with Cure53, they know their stuff. The report largely speaks for itself in conveying their level savvy on browser security. Google didn't hire some corporate consulting firm that would give them a glowing recommendation based on a review by folks not knowledgeable enough to really be able to differentiate between browser security architectures.

P.S. We (GitHub) have engaged Cure53 several times, including an assessment dedicated to Content Security Policy bypasses across various browser implementations. Mario (and team) are incredible to collaborate with.

ptoomey3··on Discontinue support for weak cryptographic standards
Just a quick preview...we will be posting a blog post tomorrow morning with an update. Because of the delayed final blog post, we are delaying things by one week. The plan is to do a 1hr brownout on February 8th, followed by a permanent deployment on February 22nd. We realize that no matter how much notice (announced roughly a year ago), there are bound to be some systems that are both incompatible and unknowingly running on production systems. So, the hope is that a 1hr brownout will help to unearth such systems and give folks time to get them updated.
ptoomey3··on CircleCI trusts 8 analytics companies with your source code and API tokens
You basically report the data to your own servers and then relay whatever subset you like to google using their “measurement protocol”: https://developers.google.com/analytics/devguides/collection...
ptoomey3··on CircleCI trusts 8 analytics companies with your source code and API tokens
Thank for the kind words. As Neil noted below, we would like to lock things down even further by proxing all data that is sent to Google Analytics. And, as a bonus, it would remove the last destination host for content exfil attacks that we know of. Our strict CSP policy has been a nice win, but the strictness has made it that much more clear that allowing nearly any third party sites, even for innocuous things such as images/xhr, isn't ideal. And, we are always on the lookout for more bypasses: https://bounty.github.com/targets/csp.html.
ptoomey3··on Crypto Anchors: Exfiltration Resistant Infrastructure
We recently developed/deployed a simple “crypto as a service” API for other apps/services to use for easy encrypting, integrity protecting, etc. It was originally developed with an HSM and eventually decided to redo it without. There were lots of unanswered questions with the HSM in terms of having the operational experience to know how they would scale across data centers, how well replication would work, how well failovers would work, etc. We had much stronger confidence in a plain old golang service, MySQL, and leveraging Vault as a master key issuer. We basically key wrap/integrity protect everything in the DB and present a simple Grpc interface. An HSM would have been nice, but a small/simple service isolated from other systems largely gets us what we want, and with the confidence to scale it as we would any other application.
ptoomey3··on Zero-knowledge proofs, Zcash, and Ethereum
You can frame the questions in a way that discloses no information to the prover. The questions ideally would have a 50% guess rate either way and leak no info. So, a question like “is her name Alice or Eve?” is a bad one, as you have leaked that her name is one or the other. But, questions like “is her birthday in the first half of the year or second” or “is her birthday on an even day or odd” (modulo there being 31 days in some months) leak no information if you don’t tell them if their answer was right or wrong. You come up with 20 such questions and then give them a pass/fail. No information is leaked, but the prover has a very low chance of guessing 20 50% chance questions correctly in a row.
ptoomey3··on Se­cu­rity Keys
SoftU2F was designed fully conscious of this. It was largely developed to help push forward the idea of a software based token with every hope that, over time, native browser and operating system implementations push things fully into the mainstream.
ptoomey3··on Passwords Evolved: Authentication Guidance for the Modern Era
For sure, software based tokens are not the norm (yet). Part of our hope with releasing a software based token is to push the security vs. user experience conversation forward. Sure, hardware tokens are great. But, they don't have the best UX and have a pretty steep barrier to entry. Hopefully Soft U2F helps to spawn a more varied set of solutions. Over time, browsers and password managers could integrate these kinds of features. For example, https://crbug.com/678128 is a tracking issue for implementing U2F in Chrome itself, which would be great.
ptoomey3··on Passwords Evolved: Authentication Guidance for the Modern Era
You aren't the first: https://twitter.com/search?f=tweets&q=patricktoomey%20sentoo... :-)
ptoomey3··on Passwords Evolved: Authentication Guidance for the Modern Era
There is no reason software based U2F tokens can't exist. We actually released one just this week: https://github.com/github/softu2f. It went through a certification process and was certified as a valid FIDO U2F implementation. I'd love to see more and more non-dongle based solutions going forward.
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
Once you have malicious software running it is largely game over. Sure, the hardware token can require a press..but once pressed what challenge is being signed? Malware can just wait and send a challenge for Site A when you are actually trying to sign into site B. Or, the malware can just wait until you login and steal your browser cookies. Oh, also, Soft U2F can require a similar physical touch if you have a mac with Touch ID.
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
The primary reason is exactly the reason you cited (u2f support is not ubiquitous across browsers..especially mobile). We may consider allowing folks to use u2f exclusively in the future, but we started conservatively given the already risky proposition of account lockout with regular 2FA.
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
Jinx :-)
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
What is the attack scenario you feel a hardware token protects you against that a software token will not (for the use cases U2F was designed for)? Sure, hardware tokens prevent malware from actually lifting your private keys. But, to steal your software private keys you likely need malicious code running on your computer. And, once an attacker has that, it is largely game over for all intents and purposes anyway. They can ask your hardware token to sign bogus requests, steal your passwords, etc. Sure, with a hardware token you can wipe your machine and feel semi-confident that you get to keep your private keys. But, really, once your machine has been compromised and you wipe it, setting up new private keys sounds like a wise practice regardless. I'm not arguing that hardware tokens have zero use. But, for most users, the attack model where hardware tokens shine is likely not of value to them.
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
And the same is true on GitHub. You can use app based TOTP without SMS.
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
That is roughly all U2F is. It is a per-origin key pair that is registered with each site and used to sign challenges. At some point browsers themselves might implement something like Soft U2F, at which point, they basically will have "improved the UX of client certs".
ptoomey3··on Soft U2F: A software-based U2F authenticator for macOS
Mostly because, thus far, it has the best user setup experience (doesn't require user to download a new app and hence can often get enabled in seconds) and has the best "lost device"/"broken device" story (people tend to not have their backup codes). I think things like Soft U2F can change that equation a bit. An iCloud keychain synced 2FA credential would go a long way toward addressing some of the usability issues with traditional TOTP based 2FA.
ptoomey3··on GitHub's post-CSP journey
Indeed, lcamtuf's article was a reference during this effort and was linked to in our CSP post last year: https://githubengineering.com/githubs-csp-journey/
ptoomey3··on GitHub's post-CSP journey
I've updated the post to spell out/link to a description of CSP upon first mention.
ptoomey3··on GitHub's post-CSP journey
I have updated the article to link out and define the first reference of CSP.
ptoomey3··on GitHub's post-CSP journey
Yeah, the missing context is that we are talking about vectors where GitHub would not be sanitizing the input correctly. In other words, vectors that traditionally resulted in XSS. We do exactly as you say in places where we expect user controlled input. For example, all issue/pull request comments are Markdown. And, for security, we go to great lengths to ensure that we only accept a subset of HTML that is safe AND that the resulting HTML is well formed. But, as history has shown, XSS is more or less unavoidable. There are just too many places where it can occur for any application to 100% avoid it. This is at the heart of CSP. Given that history has shown that XSS was unavoidable, the idea was to add a browser feature as a second line of defense. So, the article is written from the perspective that traditional XSS is neutered (the whole "scripting" bit of XSS is gone) using CSP. So, given that, what might an attacker be able to do without injecting any JavaScript? This is the origin of "scriptless attacks". Dangling markup is the most popular technique for exploiting a scriptless attack. So, it isn't that GitHub would be failing to create well formed HTML. It would be a scenario where an attacker would traditionally like to have injected a `<script>` tag, but is no longer able, so they must go to the next best thing...dangling markup.
ptoomey3··on GitHub's post-CSP journey
Yes, the attack assumes a content injection bug in GitHub.com. The attack is not using our own gravatar URL generation against us; it is the attacker crafting an arbitrary URL and using that URL inside of an arbitrary image tag. The reason for the attacker being "forced" to use a gravatar URL is that it was one of the very few third-party hosts we previously allowed by our CSP policy. So, the attack demonstrates how this previously allowed host could be used to exfiltrate sensitive content if/when an attacker found a way to inject arbitrary HTML into a page on GitHub.com.
← PreviousPage 2 of 3Next →