[tex-live] TL 13 -- how many maintained installations are there?

Zdenek Wagner zdenek.wagner at gmail.com
Wed Apr 16 11:28:34 CEST 2014


2014-04-16 11:05 GMT+02:00 Philip Taylor <P.Taylor at rhul.ac.uk>:
>
>
> Winston Smith wrote:
>
>> Since when did it become acceptable to
>> a.) introduce phone-home mechanisms into open-source software, and
>
>
> The fact that TeX Live is so-called "open source" is completely
> irrelevant to the discussion; we are discussing whether and how
> it might be augmented to collect potentially useful statistical
> information, which is a technical/philosophical issue rather than
> a political one.
>
>
>> b.) defame the personal decision of a software maintainer against this
>>       as dictatorship, and
>
>
> It is not the decision itself, but rather the fact that, according to
> Reinhard, what can or cannot go into TeX Live hinges solely on the
> opinions of one person.  If that is not a dictatorship, what is ?
>
>
>> c.) accept surveillance and erode privacy (how many TeX-users will be
>>        aware of that new "feature" and know how to turn it off?) under the
>>        premise that surveillance is in place everywhere else (e.g. the
>> phone
>>        example)?
>
>
> No-one is being asked to accept it; we are mooting the idea, and
> discussing whether or not it might be worth implementing.  Only
> when that decision has been reached, and only if the decision is
> "yes, it is technically feasible and worthwhile", need we worry
> about how to inform the users of the fact.  And if we /do/ decide
> that the idea is worth implementing, then informing the users will
> be a piece of cake, since when TeX Live is installed it can inform
> the user that, unless they select the "opt out" option, the TeX Live
> manager will henceforth collect anonymised usage statistics and report
> them to a central repository.
>
Cetral user tracking is not a question of living with or without
guilt, it is the matter of what is done with TeX. anonymising never
works. You always get the report from a known IP address. You can
therefore make statistics per IP and find whether it is a fixed IP
address or whether it is a public IP address of NAT having several
computers behind or whether it is an address obtained from DHCP.
Mathematics and the properties of the TCP/IP protocol offers you
enough tools to do it. Thus the user may be tracked not only in the
central repository but even on many routers. It need not be a problem
of a user who creates free documents only. However, research is often
secret until it is published. I am bound by contracts not to reveal
the results of my research. In this case being tracked equals to the
life with guilt. Even Czech National Bank uses LaTeX because the
admins found that TeX is safe and does not report any potentially
sensitive information anywhere. Implementing any kind of user tracking
may result in placing TeX to a blacklist in many organizations.

Personally, I would opt-in to report how many installatins I have on
which platforms but not more. I am not allowed to report more.
>
>> The whole idea of phoning home goes against half a dozen principles
>> of open-source software. There are certain things you will never see in
>> good (!) open-source software. Phoning home is one of them.
>
>
> As I wrote above, the fact that TeX Live is so-called "open source"
> software is completely irrelevant to the discussion.  I use TeX Live
> because it provides (some of) the functionality that I need, just
> as I use Microsoft Windows, Adobe Acrobat, Mozilla Seamonkey and
> Easeus Backup.  Three of those five are not open source, and that
> fact is just as irrelevant to my choice to use them as the fact
> that TeX Live and Mozilla Seamonkey /are/ "open source".  The
> choice of what software to use should be made solely on the
> grounds of functionality and cost, not on overtly political criteria
> such as the putative benefits of free-as-in-libre/open-source software.
>
>
>> Naming and shaming does no-one any good.
>
>
> On the contrary, if one does not "name and shame", all manner
> of heinous things can be (and often are) conveniently swept under
> the carpet.
>
> Philip Taylor
>



-- 
Zdeněk Wagner
http://hroch486.icpf.cas.cz/wagner/
http://icebearsoft.euweb.cz




More information about the tex-live mailing list