wpeterson

9 ms·

31.
by wpeterson 11y ago
Perhaps a better way of framing this is: Distributed repos make writing software easier. Monolithic repos make testing and running software easier. Large monolithic repos are an artifact of scale where the individual cost of lost productivi
32.
by wpeterson 11y ago
Architecture is just dividing pain into different buckets. Large repositories are painful, as each developer bears the pain of integration every time they have to make a change to the repository integrating with all of the other code. Small
33.
by wpeterson 11y ago
The weather is awesome and it's a beautiful area. However it's more crowded and expensive than many cities without any of the benefits of living in a proper city. The housing and services are throttled despite an ever increasing
34.
by wpeterson 11y ago
+1 for working remotely in Mountain View. As a new resident, I wish I were living somewhere else and working remotely.
35.
by wpeterson 11y ago
If this system is merely decoding JSON and writing payloads to S3 for asynchronous data processing, why not have your clients write directly to S3?
36.
by wpeterson 11y ago
Did someone say Gavin Belson and Nucleus?
37.
by wpeterson 11y ago
For the sake of error handling on rollback, you're usually better delayed job enqueueing events to after commit hooks if you're concerned about failures resulting in a rollback.
38.
by wpeterson 11y ago
This was a well written article with an interesting investigation. However, storing small, ephemeral messages like jobs in a queue within Postgres is a bad idea and the pain far outweighs the benefits of transactional rollback for jobs. Ins
39.
by wpeterson 11y ago
Needs more cowbell!!!! 3 is not enough.
40.
by wpeterson 12y ago
Hermit crabs are great scavengers and mostly peaceful, but they hardly wait for a shell to be vacant before moving in. They will definitely kill a live snail to take their shell if they need a bigger home. Or another crab.
41.
by wpeterson 12y ago
There's interesting social psychology to back that up: status ambiguity as the fastest way to endear yourself to the new group. Representing higher status than the entire group marks you as a threat. Representing lower status than the
42.
by wpeterson 12y ago
It looks like you didn't allow for using more than a single process by setting GOMAXPROCS. Additionally, it looks like you're re-seeding your random engine inside your sample loop, which is very slow. You only need to seed the eng
43.
by wpeterson 12y ago
Is this a game purely for entertainment or a vehicle for gathering mobile data?
44.
by wpeterson 12y ago
Agreed on issuing new shares for options pools. This dilutes all pre-existing equity holders equally.
45.
by wpeterson 13y ago
There are open support tickets in both Heroku and AWS about the impact of this bug but no answers yet. I hope folks will promote a warning if either platform is effected both on HackerNews and twitter.
46.
by wpeterson 13y ago
Hi, I'm an engineer building Polar Polls. This is an example of one of our realtime poll results dashboards. We've built some cool tools for building beautiful, photo polls and the ability to embed them anywhere on the web. http
47.
Could This App Replace Commenting On Mobile Devices?(fastcodesign.com)
1 points by wpeterson 13y ago|0 comments
48.
by wpeterson 13y ago
I'm impressed at the closure based pagination as a novel way to solve a problem. Unfortunately, the user experience is quite unpleasant. I hit expired pagination links several times a day. You guys can do better.
49.
by wpeterson 13y ago
Rap is music, hip-hop is culture.
50.
by wpeterson 13y ago
I could not disagree with this more. Focus on building your business and your product, not devops infrastructure. Unless your business is devops infrastructure. Your initial technology stack should allow you to deploy code seemlessly from
51.
by wpeterson 13y ago
We had the same experience: CloudFront to Fastly. Fastly has a great CDN service, beutiful origin hosting for a Rails or Python app, and customizable output filters (gzip? yes, please) and customizable cache config. Can't recommend the
52.
by wpeterson 13y ago
It looks like our data on Polar Polls predicted the new logo selection in advance: http://polarb.com/polls/tags/yahoo30
53.
Are Jobs Obsolete? (2011)(cnn.com)
2 points by wpeterson 13y ago|0 comments
54.
Microsoft And Polar Almost Built A Killer Smart TV(fastcolabs.com)
2 points by wpeterson 13y ago|0 comments
55.
by wpeterson 13y ago
This assumes you don't write any tests. In that case your system has more problems than the scope of this article. There's nothing wrong with using metaprogramming to generate methods, as long as you are writing tests for those m
56.
by wpeterson 13y ago
I'm surprised no one mentioned the Rainbows app server, a multi-threaded version of unicorn. Unless you're running jRuby, you're limited by Ruby's green threads. If you want to saturate all of the cores on your server, y
57.
by wpeterson 13y ago
This has worked well for our teams. As a social norm of software development I always IM/email someone with their headphones on and don't interrupt them. It's also a nice courtesy to IM someone before dropping by their desk so they aren't
58.
by wpeterson 13y ago
I worked in a similar, one-giant-open-area office for a year and didn't enjoy it. My team was a cluster of 8 fixed desks in a giant sea of engineering, with people always walking through our area on their way somewhere else. The noise leve
59.
by wpeterson 13y ago
Separating focused, creative work from interrupt-driven work is essential. Context switching between them is expensive. On my engineering team we tried to collect all interrupt driven work together (triaging bug reports, working on deploym
60.
by wpeterson 13y ago
I agree with having a variety of public spaces (conference rooms, lounges) for the right context. I've also found that open plan is great up until you hit the size of your team. One large, open room with only people you work directly with

More ›