Date: Sun, 11 Sep 2011 14:35:47 -0600 (MDT) From: Warren Block <wblock@wonkity.com> To: Chris Rees <crees@freebsd.org> Cc: freebsd-ports@freebsd.org, Greg Byshenk <freebsd@byshenk.net> Subject: Re: Removed ports - looking from the bench Message-ID: <alpine.BSF.2.00.1109111316010.10961@wonkity.com> In-Reply-To: <CADLo839ohQP=Rud4st_02%2Bx_-aJo7KTx-0ZKBVVoyQx1XUh2Rw@mail.gmail.com> References: <4E6B1AF5.7090900@tomse.dk> <4E6B227B.5050708@FreeBSD.org> <4E6B77EE.6030509@gmx.de> <20110910171530.GC23457@guilt.hydra> <CADLo83_jp3Orrjp0LvVRnwxY-RL=0e8PSXmEruwaH7TsLZ7=VA@mail.gmail.com> <20110910190549.GA23971@guilt.hydra> <20110911124910.GK13219@portland.byshenk.net> <alpine.BSF.2.00.1109110739170.10077@wonkity.com> <CADLo839ohQP=Rud4st_02%2Bx_-aJo7KTx-0ZKBVVoyQx1XUh2Rw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
---902635197-1833353319-1315773347=:10961
Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8BIT
On Sun, 11 Sep 2011, Chris Rees wrote:
> On 11 September 2011 15:35, Warren Block <wblock@wonkity.com> wrote:
>> On Sun, 11 Sep 2011, Greg Byshenk wrote:
>>>
>>> On Sat, Sep 10, 2011 at 01:05:49PM -0600, Chad Perrin wrote:
>>>>
>>>> Why?
>>>
>>> Because, in the cases here under discussion, there is somethin "wrong"
>>> (for some value of 'wrong') with the software in question. I can't
>>> speak for Matthias or Chris, but I think the point here is that (at
>>> least some) people don't want to make foot-shooting easier.
>>
>> Slippery slope: consider PHP, or Apache, or any MTA. Or newfs.
>
> No. PHP, Apache and the MTAs are maintained. Newfs is not buggy.
There is "something "wrong" (for some value of 'wrong')" with all of
these. newfs will easily overwrite an existing filesystem, for example.
That's the slope, the degree to which ports or FreeBSD is going to go to
assume ignorance on the part of the user and protect them from
themselves. Historical precedent is to inform the user about problems
but otherwise assume they know what they're doing. Certainly that's
wrong at times, but the other way is the road to "That's dangerous and
therefore not allowed."
Whether there's overt questioning or security through obscurity, there
is no way for the software to take on the responsibilities of the
operator.
> Straw men are the tool of someone who has no more valid points to
> make, remember that.
As is calling a point a straw man rather than addressing it. Probably
neither is quite right.
Let me suggest a reasonable[1] plan:
Modify portdowngrade[2] or create another tool[2] to retrieve removed
ports. Show the scary reason for removal before getting files from CVS.
The user acknowledges that implicitly by retrieving files, or explicitly
by answering an "Are you really, really, ultra-double sure?" question.
The existence of this tool satisfies[3] users who want to install old
ports.
Continue the removal of dead ports as it has been going.
If archival of old historical distfiles is needed, that's not really a
FreeBSD problem. Start a new project with its own website. Quick,
somebody register deadports.org![4]
[1] All reasonableness is subjective.
[2] Not it!
[3] Well, no, some people won't be satisfied, ever. But this would
address the problem and might mollify or assuage or assistify.
[4] There may be something out there already that can be used. In fact,
I think we'd all be surprised if there wasn't.
---902635197-1833353319-1315773347=:10961--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1109111316010.10961>