Live data from Hacker News

Dear GitHub

github.com

271–280 of 491 posts

Re: Dear GitHub

#271

Hi Adam, Addy, Andreas, Ariya, Forbes, James, Henry, John-David, Juriy , Ken, Nicholas, Pascal, Sam, Sindre, My name is Jono and I started as Director of Community back in November at GitHub. Obviously I am pretty new at GitHub, but I thought I would weigh in. Firstly, thanks for your feedback. I think it is essential that GitHub always has a good sense of not just what works well for our users, but also where the pa…

Nice to meet you Jono. I maintain a similar list, albeit a bit more focused on GHE. https://github.com/kevinSuttle/github/issues

Re: Dear GitHub

#272
post #270
post #141

We need world class, modern, distributed bug tracking now. If you google around for this technology, a lot of nice ideas, many using git itself as transport, were poking around, and around 2009 they started falling silent. Why? Because GitHub started up and everyone just buzzed over to it like so many moths to a flame, having learned nothing from places like Sourceforge about what happens when 90% of the open source…

You mean like GitLab?

GitLab is still centralised (whomever is hosting…), not distributed.

Re: Dear GitHub

#273
post #78

Earlier quoted context omitted.

There's an old story about this man who stood quietly next to a closed door in Moscow, said nothing to no one, and did nothing else of interest. Eventually others joined him, and before long a queue has formed. No one knew what they were standing in line for. Monkey see - monkey do.

One of Milgram's experiments (not the infamous one) tested this, using people standing on the sidewalk looking up. https://youtu.be/P0e6zG8IbE8

I really want to try this out myself now.

Is it really possible that people just continue standing there while not getting why they are doing it? Don't they even care to ask?

Feels like it would take just a single guy to ask: "What are we looking at?" For everyone to realize that they are acting like fools.

Re: Dear GitHub

#274
post #78

Earlier quoted context omitted.

There's an old story about this man who stood quietly next to a closed door in Moscow, said nothing to no one, and did nothing else of interest. Eventually others joined him, and before long a queue has formed. No one knew what they were standing in line for. Monkey see - monkey do.

One of Milgram's experiments (not the infamous one) tested this, using people standing on the sidewalk looking up. https://youtu.be/P0e6zG8IbE8

I really want to try this out myself now.

Is it really possible that people just continue standing there while not getting why they are doing it? Don't they even care to ask?

Feels like it would take just a single guy to ask: "What are we looking at?" For everyone to realize that they are acting like fools.

Re: Dear GitHub

#275

Earlier quoted context omitted.

Recently even Python language planned to move its repositories to github.com for network effect, instead of helping projects like kallithea SCM and trac by partnering with software conservancy or gnu. Python should learn a lesson when they decided to move their repository to closed source system like github. But obviously as people use Facebook, developers use github for the same reason, network effect. I hope they c…

> instead of helping projects This isn't how it works. You don't help projects by pretending the football-stadium-sized issues with them don't exist and using them despite their flaws. Trac is an awful, awful piece of software. It's awful to set up, to use, to maintain, to gather feedback from, it's awful for just about everything. If in some very weird parallel universe it gathered even 1% of the following that Gith…

Ubuntu has actively moved away from directing users to Launchpad to report a bug from a crash, and instead moved towards reporting to a crash database where they're triaged by actual developers.

Debian, in a similar vein, doesn't even have a web form to report bugs. As a result, almost every bug I've gotten on a Debian package has been clueful. At the very least, I know what version of the software they're running.

I've managed repositories that have gotten hundreds of low-quality issues. The poor filtering and curation functionality of GitHub ensures that a good deal of my spare time I have to work on those projects is spent managing the inbound issue flow, which is at least 90% noise.

Re: Dear GitHub

#276
post #268
post #173

The author mentions that if GitHub was open source, they would implement these features themselves. Gitlab[1] is an open source repository manager that supports local installs as well as public hosting at gitlab.com. If author appreciates open source, perhaps they should put their efforts into improving an existing open source option rather than relying on a proprietary solution. [1] https://gitlab.com/gitlab-org/git…

Absolutely agree! I cannot fathom why people are still actively supporting GitHub. Even if you ignore the ethical reasons, which if you are an open source developer really should suffice, GitLab is better and more customizable in every way. Supporting it benefits yourself and all of the FOSS community.

gitlab.com is dog slow and my code is too important to risk self hosting.

Re: Dear GitHub

#277

It's 2016, and GitHub is stagnant. GitHub used to bill itself as "Social Coding", but the "Network" graph has not seen ANY updates since its original introduction in April of 2008 . Issues has seen very few updates. Even the OSS projects that GitHub uses internally have grown stagnant as GitHub runs on private, internal forks and maintainership passes to non-GitHub-employed individuals (e.g. https://github.com/resque…

> They're chasing some other goal...whatever it is. I've been puzzled for a while with what github is doing hiring so many social impact employees. https://twitter.com/agelender https://twitter.com/_danilo https://twitter.com/rachelmyers https://twitter.com/nmsanchez https://twitter.com/BiancaCreating https://twitter.com/ammeep https://twitter.com/davystevenson Maybe something more noble than a social coding site?

  Maybe something more noble than a social coding site?
I doubt it. Github has a reputation problem. I wouldn't put anything sensitive on there, given the attitude github leadership showed about privacy ethics in the Julie Horvath incident.

Re: Dear GitHub

#278
post #241

Earlier quoted context omitted.

One of Milgram's experiments (not the infamous one) tested this, using people standing on the sidewalk looking up. https://youtu.be/P0e6zG8IbE8

He might think he's testing conformance, but he actually just tested curiosity.

Wouldn't they at least ask after a while if that was the case?

Re: Dear GitHub

#280
At Sourcegraph, we're trying to help solve these problems for developers everywhere (https://sourcegraph.com), both in open source and inside companies. GitHub’s commercial success and contributions to the world of development are impressive (and I'm speaking as a GitHub user for 8 years), but they can’t build everything developers need on their own.

We’re really pumped about improving dev team collaboration in the GitHub ecosystem by (soon) letting anyone use Sourcegraph.com’s code intelligence (semantic search/browsing), improved pull requests, flexible issue tracking with Emoji reactions instead of +1s (example: https://src.sourcegraph.com/sourcegraph/.tracker/151), etc.—all on their existing GitHub.com repositories.

All of Sourcegraph’s source code is public and hackable at https://src.sourcegraph.com/sourcegraph, so it can grow over time to solve the changing needs of these projects. (It’s licensed as Fair Source (https://fair.io), not closed source like GitHub or open source.)

Email me (sqs@sourcegraph.com) if you’re interested in beta-testing this on your GitHub.com repositories.

Post reply on HN