Silly time stamps

Hi,

This is an example of timestamps are in git:
    tree c67dc912a777ef6198a5c76890ebf11cd8ccc625
    parent 6e527d8e07a80058bf8ac46180e25a9d4cb745eb
    author my name <m@test.org> 1397073956 +0400
    committer my name <m@test.org> 1397073956 +0400

    testing again

As you can see, it presents the timestamp in UNIX format (I suppose it's UTC) plus an additional information which is an offset that describes my time zone.

Now, the way UTC works is that every client is supposed to apply his own offset on it (based on locale stuff) so that the user-interface (let it be CLI or GUI) can print an pretty time format based on his location.

The offset that is used is local to each client. E.g. if it's UTC 10:00, it will look 14:00 (cause +04:00) in my part of the world, while looking as 12:00 in another part of the world (+02:00 time-zone). This time-zone offset is local to the computer based on locale stuff. The only thing that we need is the UTF time stamp (and then anyone can make it look pretty based on where he is living).

Now, the problem is that git shares the UTC time (totally OK) AND my fucking time-zone! Why would anyone give a damn rat's ass about my time-zone? What difference could it make if you know that the clock looked like "14:00" on my part of the world?

This was raised previously[1], but none of the responses are convincing.

One guy was saying here[2] some nonesense like "I'd prefer not to lose the information. If someone has committed a change at 2am, I like to know that it was 2am for _them_. It helps me decide where to look first for the cause of problems. :)"

How about we also modify the time data to augment it with my mood? E.g. "1397073956 +0400 totally_pissed_off".

The same asshole in [2] gave another retarded example that is just less visible to non-humans: "It also helps disambiguate certain comments, especially those involving words or phrases such as 'yesterday' or 'this afternoon'.".

This is equal to say: instead of curing the root cause of the problem (by educating assholes to not use ambiguous time information like "yesterday"), let's fuck the software instead (by encoding useless information here and there).

[1] http://thread.gmane.org/gmane.comp.version-control.git/52 [2] http://article.gmane.org/gmane.comp.version-control.git/114

On 04/09/2014 10:50 PM, Mahmoud Asshole wrote:

Please conduct your discussions here in a civil tone. It is both more pleasant for all involved and also more likely to elicit a response. I hardly think that the "waste" of 12 bytes in every commit is an act of stupidity so inexcusable that it would deserve your bile, even *if* one were to agree that this information is useless (which I personally don't think).

Thanks, Michael

Am 11.04.2014 13:08, schrieb Michael Haggerty:

I would guess he is more concerned about the unnecessary disclosure of information that could be used to track or (together with other data) identify you.

Since the reasons to include it seem to be specifically to know more about the comitter this seems to me the typical conflict between privacy and security.

On 11.04.2014, at 15:19, Holger Hellmuth <hellmuth@ira.uka.de> wrote:

Perhaps, though I fully agree with Michael that calling people "assholes" because their opinion differs from yours is not increasing the chances that anybody will listen to you, or bother to try and understand your problem.

More between "privacy" (or perhaps "personal safety"? think: dissident coder?)  vs. "feature that is useful to some people".

If one is truly concerned about spilling timezone information, how about this:

  alias git='TZ=0 git'

Seems to work for me, at least in bash.

Cheers, Max

Am 11.04.2014 17:14, schrieb Max Horn:

Well, at least the reason mentioned in the gmane citation about knowing if it was 2 am for them, is strange. Did anyone ever check this timestamp to see if a patch was made at an unusal hour? What does it tell you? Some people sleep from 4am to 11am and do their best work at 2am. Only if you know someone you could infer he made a patch at an unusual time, but then you probably also know where he lives and have this information anyway. Your best and only indication of a sub-par patch is the patch itself.

The only reason I can see is as a warning signal or indication that a patch wasn't sent by the person whos name is on the cover.

From: "Holger Hellmuth" <hellmuth@ira.uka.de>
To: "Michael Haggerty" <mhagger@alum.mit.edu>
Cc: "Mahmoud Asshole" <a1209145@drdrb.net>; <git@vger.kernel.org>

If you look at "Mahmoud's" email address, it is from http://10minutemail.com/ 'the best disposable e-mail service.'.

So it looks like s/he knew what they were doing when venting their spleen.

It would have been nicer to know the underlying source of their frustration, rather than them going off the rails.

Philip

On Wed, 09 Apr 2014 20:50:57 +0000, Mahmoud Asshole wrote: ...

Please go directly to the special hell of debugging timezone-related stuff in customer data, for which this extra bit of information is heaven-sent. Do not collect additional bile, and do not reappear here.

Andreas