Earlier quoted context omitted.
You're able to add +1 through their recent GitHub Reactions update now: https://github.com/blog/2119-add-reactions-to-pull-requests-...
Except it doesn't do notifications like a comment does...
Good.
Saved Replies
81–90 of 91 posts
Re: Saved Replies
#82Bad maintainers will rejoice at the possibility of replicating a fiasco like this: https://github.com/ansible/ansible-modules-core/pull/2023 with zero effort. Saved replies are a silver bullet only for those who already took the time to compose meaningful (engaging?) replies. For all the others, a quick way to shoot themselves in their feet.
Re: Saved Replies
#83This is awesome. I have a repo with only about 1k stars, so nowhere near the adoption of something like npm, etc. Even on my repo, I still regularly receive issues that say "it doesn't work. Why?" With no reproducable info. These will cut the time on asking for that info and give more time back to other users or the project itself.
Don't forget https://github.com/blog/2111-issue-and-pull-request-template... - even for private repos it really has changed the way we do bug reporting (for the better).
I use this on Cachet [1] and find that most of the time people just remove the pre-filled content regardless.
Re: Saved Replies
#84Bad maintainers will rejoice at the possibility of replicating a fiasco like this: https://github.com/ansible/ansible-modules-core/pull/2023 with zero effort. Saved replies are a silver bullet only for those who already took the time to compose meaningful (engaging?) replies. For all the others, a quick way to shoot themselves in their feet.
That's gotta be some automated thing gone bad though. Surely a human would not be conducting a conversation like that.
Re: Saved Replies
#85Earlier quoted context omitted.
I would be okay with these responses if it was remotely apparent whether Issues acts as a discussion forum or not. It varies for every project unfortunately. For smaller projects, the risk of asking on SO is the maintainer, and anyone else familiar with the project, will never see it as they wouldn't be actively monitoring it. Hopefully Github's current blast of Issues updates will make it easier for project owners t…
This is where CONTRIBUTING.md comes in. If a project doesn't want Issues to be used for support, then it should mention it there. Whether people actually read it is another story :(
It's not at all logical to look into the guidelines for contributing to a project of what you seek is support, not to contribute a feature request or a patch. I'd never look there.
Re: Saved Replies
#86Earlier quoted context omitted.
The notion of shortcuts for inserting prepared text isn't exactly new. I was working in a data entry job in 1983 and the system would expand predefined text sequences into the matching full text as needed. Doing this through a pretty dropdown is the same idea. I'll bet the idea wasn't new in 1983 either.
The implication here, isn't so much that what was created was new, it's "Would this feature had been created had somebody else not shown GitHub a better way?" Since the commit history for https://github.com/notwaldorf/github-canned-responses/commit... is fairly new, it's easy to believe this feature was in GitHub's pipeline for a while. If this wasn't the case, it can sting from a brand perception point of view. GitH…
> GitHub is suppose to be synonymous with innovation...
What was their innovation previously, though? I think they built a pretty great site around git, but I struggle to see where this massive innovation is.
Re: Saved Replies
#87Earlier quoted context omitted.
Hi andrewstuart'); DROP TABLE repositories;--, Thanks for sending this in! Based on my reading, this is working as intended and is not a bug in the code. Let me know if I'm misreading something. Thanks again!
That's the output, yes, assuming proper input sanitization, which I would not expect to be a problem given that it's github and they deal with a ton of the most dangerous kind of dynamic content (any kind of code) already. It was a bad joke, I guess.
"Sanitization" is the wrong approach. I would hope Github handles things right, by keeping track of what every value represents.
Re: Saved Replies
#88Check out their documentation for less details on the feature: https://help.github.com/articles/working-with-saved-replies/
Re: Saved Replies
#89Earlier quoted context omitted.
That's the output, yes, assuming proper input sanitization, which I would not expect to be a problem given that it's github and they deal with a ton of the most dangerous kind of dynamic content (any kind of code) already. It was a bad joke, I guess.
"Sanitization" is the wrong approach. I would hope Github handles things right, by keeping track of what every value represents.
With interpreted text languages, and indeed with machine code, there are no types to associate with a value and a value may very well be executable code. This invariably means that the only approach is to "sanitize" your output for a given context if the type associated with it means it should only be interpreted as data.
Re: Saved Replies
#90Earlier quoted context omitted.
"Sanitization" is the wrong approach. I would hope Github handles things right, by keeping track of what every value represents.
With interpreted text languages, and indeed with machine code, there are no types to associate with a value and a value may very well be executable code. This invariably means that the only approach is to "sanitize" your output for a given context if the type associated with it means it should only be interpreted as data.
Interpretation does not prevent typing, and machine code is an execution format. The right thing in any language is: express the operations that you are going to perform in a form that clearly distinguishes between values that represent different varieties of thing. (Types make this easier, otherwise you have to implement more of it yourself).
You talk about code and data as though they were the only things, but they're not; getting one variety of data when you were expecting another variety can just as easily lead to security bugs as getting code instead of data or vice versa. Sanitization very rarely works - and in the rare cases where it does, it still indicates a deficiency in the underlying model.