Has the port collection become to large to handle.

[pauls at utdallas.edu]

--On May 13, 2006 7:18:16 PM -0400 Garance A Drosihn <drosih at rpi.edu> wrote:

> At 2:28 PM -0400 5/13/06, fbsd wrote:
>> To all question list readers;
>>
>> Now with 14576 ports in the collection where do you
>> draw the line that its too large to be downloading
>> the whole collection when you just use 10 or 20 of
>> them?
>
> This is a good question.  For all those people who want
> to roll their eyes and ignore this question, please
> answer it.  Where *DO* you draw the line?  Obviously it's
> not at 10,000 ports.  Will it be 20,000?  50,000?  How
> many programs exist?  Will every single program known to
> man eventually be in the ports collection?  How hopeless
> is that?  And if not, then "Where do you draw the line?".
>
I must confess to some confusion here.  I don't have any boxes (and most of 
mine are headless servers) that have only 10 or 20 ports installed.  (Cvsup 
alone has two dependencies, so you'll have three ports installed just to 
keep ports up to date.)  Most of my servers have over 100 installed ports 
at least.  My workstation is close to 490.  Furthermore, the *entire* ports 
tree, **up-to-date**, consumes 329MB of space (without distfiles).  In the 
days when you can buy 20GB hard drives for $10 US, how can this be a 
problem?

The purpose of the ports collection is to make available to FreeBSD users 
programs that they might find useful.  The more programs there are 
available, the wider the audience that is attracted to FreeBSD. 
Furthermore, a port wouldn't even exist in FreeBSD unless at least *one* 
person was interested enough in it to become the creator and maintainer of 
that port.

The price *you* pay for keeping your ports up to date is very minor - 10 
minutes a day (max) updating through CVS and 329MB of hard drive space. 
And if you have to pay for bandwidth use, you only need to cvsup right 
before you have to install or upgrade a port.

Yet this suggestions is - lets make the system infinitely more complex so 
that a handful of people who dislike having to use that 329MB and do CVSup 
occasionally don't have to deal with it.  Does that make sense to you? 
Personally, I would rather burden that handful of people by making them do 
customization than to punish the entire community to satisfy their request.

I *do* think tracking downloads would be valuable, *if* there's a way to 
implement it and aggregate the data.  Knowing how many times a particular 
port is installed might open more than a few eyes.  The problem is, you'd 
have to have accurate stats from *every* mirror and those would have to be 
aggregated and collated.  Not a big problem, for sure, but still, more work 
for somebody who's already a volunteer.  But knowing how many times a 
port's distfile was fetched and how many times it was upgraded would be 
useful information.

The problem is, it would require some sort of "ping" to FreeBSD that 
included useful data, because, once the base port is installed (which is 
somewhere between 8 and 40K normally), the install is going to pull the 
distfile from the MASTER_SITE, which could be one of twenty or thirty 
different sites.  This "ping" could also include information such as which 
mirror sites are out of date or off-line.  I get the sense that tracking 
this would not be too hard, because the "intelligence" already exists 
within ports, but I'm not knowledgeable enough to tackle the problem.

Paul Schmehl (pauls at utdallas.edu)
Adjunct Information Security Officer
The University of Texas at Dallas
http://www.utdallas.edu/ir/security/