[coreboot] Why I don't like links in commit messages and comments ...

Felipe Sanches juca at members.fsf.org
Tue Aug 15 05:56:42 CEST 2017


Other that, the other solution I can think of would involve the creation a
coreboot-webreferences git repo (or wiki page, or something else) where
people have to make mandatory backup copies of any web-content they
reference on commit-messages and any other documentation. That solution is
certainly a bit harder to maintain, but much more robust if your goal is to
make sure the info will not be lost. There's a catch with copyright law in
that case, though... but there could arguably be a defense based on
historical preservation purposes similar to the Archive.org project.



2017-08-15 0:51 GMT-03:00 Felipe Sanches <juca at members.fsf.org>:

> It is not perfect, off course!
> But it is their stated mission to do the best for the longevity of the
> preserved data and they seem serious enough about that goal. And it is
> definitely better than the current state of not backing up web-references
> anywhere.
>
> 2017-08-15 0:40 GMT-03:00 ron minnich <rminnich at gmail.com>:
>
>> Why are you assuming that the internet archive will be here in 20 years
>> :-)
>>
>> On Mon, Aug 14, 2017 at 8:32 PM Felipe Sanches <juca at members.fsf.org>
>> wrote:
>>
>>> There's a rather simple solution to that: stipulate that any link in
>>> commit messages must be first saved on the Internet Archive's Wayback
>>> Machine. That is as easy as pasting the original URL into a text field in
>>> this page and clicking the "Save Page Now" button:
>>>
>>> https://archive.org/web/
>>>
>>> 2017-08-14 21:44 GMT-03:00 ron minnich <rminnich at gmail.com>:
>>>
>>>> This is an interesting headline:
>>>>
>>>> "I Bought a Book About the Internet From 1994 and None of the Links
>>>> Worked"
>>>>
>>>> What's interesting? That's only 5 years before linuxbios started. I can
>>>> make the totally unwarranted claim that the half life of a link is around
>>>> 12 years. I justify this claim with one bogus data point, namely, this
>>>> headline :-)
>>>>
>>>> But still, what's amazing is when I did a check, about 15 years after
>>>> linuxbios began, about 25% of the links were in fact dead. So the 12 year
>>>> number may not be great, but ... maybe it's not that bad either.
>>>>
>>>> And now you know why I don't like links in commit messages :-)
>>>>
>>>> ron
>>>>
>>>> --
>>>> coreboot mailing list: coreboot at coreboot.org
>>>> https://mail.coreboot.org/mailman/listinfo/coreboot
>>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.coreboot.org/pipermail/coreboot/attachments/20170815/660e73f2/attachment.html>


More information about the coreboot mailing list