> 1. I understand the thinking behind it, however I believe that,
> looking at a computer's filesystem at an overall level, it
> accomplishes very little. Most everything on a computer these days is
> NOT random. Perhaps Mozilla can be seen to "lead the charge" towards
> a fully randomized filesystem, but they should at least allow people
> who care about this to be able to turn it off.
Mozilla is not leading a charge anywhere. Past experience shows that a
great deal of nasty web-related browser exploits and so on rely on the
attacker knowing the profile directory on the local system of the user (as
many users do not change from the defaults.) This is merely a contained
and sensible response to severely limit that threat.
> 3. Personally, I simply don't like specifying a set directory during
> the profile creation process and having the program tack on some
> random subdirectory. I know that Mozilla will create certain static
> subdirectories of its own - but these names are always the same and
> can be relied upon for consistency.
This is exactly the problem :-) I remember having a similar conversation
about someone who was adamant that he wanted his legacy Mozilla profile
stored in a directory called "Profiles" and not one called "Users50". Why
does it offend your sense of aesthetics?
> 4. I can get around this annoying behaviour by copying and pasting a
> legacy profile structure that does not include the "xxx.slt"
> behaviour, then doing a search and replace of the old profile name to
> the new one. So, it's possible to "hack" Mozilla into the behaviour I
> want - it's just unpleasant and annoying to do so.
Surely a small Perl script is your friend here?
> Last week, temporarily unable to access Newsgroups, I sent a
> message experessing some of the above to one of the Bugzilla
> contributors - one of the people who, in the first bugzilla link
> above, had actually stated that they thought there SHOULD be a way of
> overriding the salting of directories. What followed was a somewhat
> confusing and frustrating email exchange that got me nowhere.
>
> One sentiment expressed was that if I wanted things to change, I
> could pay this person's company to do so.
You are, of course, referring to Ben Bucksch, the owner of Beonex, whose
business it is making improvements to Mozilla for money, and supporting
the resultant software.
> 1. It cannot be the case that when anybody in the world wants to see
> something different about Mozilla all they have to do is pay money to
> one of the people on the development team, or to that person's
> "real-life" company for which they work when they are not contributing
> to Mozilla on their own time. Not only would this raise some serious
> business practice issues, but it would also lead to various forms of
> anarchy.
Yes and no. It's not _all_ you have to do if you want to see your change
in the mainline code. As well as paying the money, you also have to
convince mozilla.org (or the relevant module owner) that the change is a
good idea. But, if you are happy with a custom version then yes, that's
all there is to it. Simple, huh? :-)
> 2. Surely, there is a group of people who are governing the direction
> which Mozilla is taking. Any suggestions for improvement would have
> to be vetted by at least a majority of this group prior to accepting
> it as a valid change to the code. I was taken by surprise by this
> person's stance which seemed to go against that grain. Had they said,
> "Thank you for your suggestion," and left it at that I would have had
> no further interest in pursuing reporting my issue with salting -
> instead they, in a round about way, suggested that my "wish list" item
> could be implemented if I paid his company money.
And so it could. He would produce a custom version of Mozilla for you with
the behaviour you wanted. If this behaviour is useful to you, this does
not seem unreasonable.
> This doesn't speak
> well for the Mozilla community at large, which surely is an open
> source community effort based on common feedback, agreement, and an
> idealistic conception of software development rather than individual,
> and capitalistic, "rogue" membership.
On the contrary. Mozilla's license was expressly designed so that
companies such as Beonex, Netscape, Intel, Nokia, Eazel etc. could take
the code and enhance it, and make money from that. Netscape adds the
proprietary AOL Instant Messenger. Nokia are doing set-top boxes with
Gecko. Beonex is concentrating on the corporate market.
Gerv
