feature request: opt-out about osrc HOT 12 CLOSED

👍, although it might make more sense to address this at the github level, API scanning permissions, etc.

from osrc.

I was actually just thinking aoubt this last night. My plan is to put a banner at the top of each page reminding recruiters, etc. not to take this too seriously and remember that it's a very one-sided view and really just meant as a toy... I've been shocked at every phase of this project by how seriously people seem to take it!

That being said, I reject your quotes around "analysis" because I'm legitimately computing stats on everyone's public GitHub activity and reporting them. I don't think that I overstate what it means at all! Remember that everything that you do publicly on GitHub is visible and I agree that you should contact GitHub support if you're really feeling uptight about that.

Either way, I'll add the option to opt-out ASAP and I'm sorry if this caused any problems.

from osrc.

P.S. do you really want to work with a recruiter that takes a stupid website like this that seriously?!? 😉

I'm mostly kidding...

from osrc.

Hi,

Before I go any further, thanks for agreeing to make the opt-out feature! I'll be happy to test it out when it's done, haha.

That said, I want to reply to what you said:

That being said, I reject your quotes around "analysis" because I'm legitimately computing stats on everyone's public GitHub activity and reporting them. I don't think that I overstate what it means at all! Remember that everything that you do publicly on GitHub is visible and I agree that you should contact GitHub support if you're really feeling uptight about that.

I'm not "feeling uptight" about my Github profile. I'm aware of the public nature of it and happy to have it. If you need me to, I'll explain why I'm upset about the analysis being done: I think it's inaccurate and misleading and characterizes me in a way that I don't appreciate. They're not even bad things, I just don't like inaccurate information about me being shown on this page.

Some specific examples from my report card; I'm sure I could find people who could easily give you similar examples:

"Casey Kolderup is an exceptional JavaScripter"

I'm not an exceptional JavaScripter. I'm actually terrible at Javascript. The reason that Github reports that I have a lot of Javascript in my repos is because I don't know what I'm doing and I have a bunch of simple, compact repos consisting of Ruby applications (on the order of ~100 lines of Ruby) that have jquery.js pulled directly into the repo, skewing Github's report of the "code" in my repo (note Github never says I wrote that code) and being interpreted inaccurately by that sentence.

"Casey is a late-week deadliner who works best around noon."

Nothing on my Github account has ever been done under "deadline." I write code, for pay, at my job, in a series of private repositories. My time spent on Github is mostly hobby work and art projects. I also DON'T work best around noon— I work TERRIBLY around noon— and it's actually hilariously evident when you look at the graph that phrase links to; zero of the actions pulled from the Github API at noon are related to coding or reviewing code or managing code. It's all starring and following repos. Which is not my work.

"These charts give significant insight Casey's character as a developer"

Grammatical error aside, it's authoritative statements like these that make this page really dangerous and counter to the "lighten up, everyone knows this is just for fun!" tone I'm apparently supposed to see. This is a public page on the internet. Tone is dangerous, and the tone this page sets is a serious one.

"Casey seems to be a pretty serious JavaScript expert with a surprisingly broad knowledge of Scala as well"

Again, totally wrong. 0% based on fact, 100% based on an incorrect conclusion using (technically true) data.

I think that what you're doing by analyzing the public activity from the Github API is really interesting but I think the tone creates a world of really dangerous possibility in misrepresenting people.

P.S. do you really want to work with a recruiter that takes a stupid website like this that seriously?!?

I don't, and that's why I'm upset— I don't want to put tools in the hands of people who don't realize that they're not helpful, because they're wasting my time by contacting them (I receive a lot of recruiter emails from people already who don't understand that I don't know Javascript and am not qualified to be, or interested in working as, a Javascript developer) and wasting their own time (because they are never going to get a response from me).

from osrc.

Damn. Sorry you feel so strongly about it. The internet is a scary place!

You can now opt-out of the OSRC: http://osrc.dfm.io/opt-out/ckolderup

Sorry again for any problems that it has caused for you and anyone else.

from osrc.

Let me know if you have any problems.

from osrc.

Thanks for taking my concern seriously and for the fast turn-around. I really appreciate it!

from osrc.

Casey, all of your points are moot. This project only pulls data publicly
visible which is exclusive of all of your private repositories. Hence the
name "open source report card" not "open and closed source report card"

On Tue, Nov 19, 2013 at 12:48 PM, Casey Kolderup
[email protected]:

Closed #56 #56.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/56
.

Richard Wilson
Sechelt Innovations
Cell - (604) 842 5318
Skype - r.crawfordwilson
Email - [email protected]

“This email may contain confidential and/or privileged information. If you
are not the intended recipient or have received this email in error, please
notify the sender immediately and destroy this email. Any unauthorized
copying, disclosure or distribution of the information contained on this
email is prohibited”.

from osrc.

@Senjai You might want to re-read the whole issue thread. In addition to the issue of (previously) not being able to opt-out, there's also the underlying issue of the data used being a representation of the repositories a user contributes to, and not a representation of the code being contributed, which will give you two massively different results. (It's why my report card said I know Python, and would soon say that I write more Python than Javascript, when I've written a total of maybe a dozen lines of Python ever)

from osrc.

@rummik Very well, but this isn't a problem with OSRC. It's simply a fun little app that aggregates public data from an api. It's also opinionated. I think it's great that this opt out functionality is integrated into OSRC, but really the issue, if there is one lies in the GH API.

From the perspective of GH, you could be able to limit public api usage if you desired. But all it would do is prevent one medium of public information from being available while the others exist (e.g. html).

I think the major gripe most users have is this:

OSRC isn't smart enough to tell what kind of programmer I am so I hate it.

and

Recruiters may use OSRC to judge my ability.

Both of these are fallacies. Firstly, osrc grabs what is provided by the GH api, it's unfeasible to go through each commit and determine what's being committed in what quantity. Secondly, anyone who thinks they should work a firm who considers OSRC a definite encapsulation of ability, or get recruited by a recruiter that thinks accordingly, is an idiot. OSRC is obviously just aggregated data. So if I write a loop of code to generate a unique html file and push that unique file 500 times each iteration, It might say I'm some sort of html freak. But there really is no context.

If anyone called me and said something along the lines of "I saw your OSRC and I was (un)impressed.." I would probably just hang.

It's a fun little application and I think @dfm did a great job with it.

from osrc.

@Senjai I'd have to agree that it's an API issue, and that @dfm did a great job on it.

For me it wasn't "I need to be careful how I market myself", so much as "I don't write Python, and now these stats are something I'm spending time thinking about". So my opt-out decision was more to keep myself in check. I mean, OSRC is cool -- and I'd love to hear what GitHub's thoughts are on expanding their API -- but I can't spend my time thinking of GitHub as a social game; which is what I wound up feeling with OSRC.

One thing that may have helped me though, is having the option to say "hey, everything I do on this repo is really this language", since the stats could then be something I could relate to.

from osrc.

@rummik Ahh, That sounds more reasonable. Not sure if it's feasible for OSRC to filter that kind of stuff.

from osrc.