Is FreeBSD just a sandbox for hackers? - BSD

This is a discussion on Is FreeBSD just a sandbox for hackers? - BSD ; In article , Michel Talon wrote: >Gnome *is* a terrible mess, from all points of view. The conventional wisdom is (or was) that Linux people were Windows users who discovered Unix and *BSD people were Unix people who discovered cheap ...

+ Reply to Thread
Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 21 to 40 of 55

Thread: Is FreeBSD just a sandbox for hackers?

  1. Re: Is FreeBSD just a sandbox for hackers?

    In article ,
    Michel Talon wrote:
    >Gnome *is* a terrible mess, from all points of view.


    The conventional wisdom is (or was) that Linux people were
    Windows users who discovered Unix and *BSD people were Unix
    people who discovered cheap hardware. I think that's
    reasonably close to the truth, or at least was for a bunch
    of years. It seems to me that that goes a long way towards
    explaining Gnome and all those other big fat messes (and the
    ports system is full of 'em). There's a reason that the
    pipes and filedescriptor abstractions exist, and it's
    definitely not to support big fat integrated one-size-fits-
    all bloatware. MH and Outlook - compare and contrast.

    >Personnally i think that one of the best features of FreeBSD, compared
    >e.g. to Open, is that it evolves quickly, and fantastic new features are
    >regularly introduced.


    I'm just coming up to speed on the new kernel and so far so
    good. It does contrast rather sharply with OpenBSD,
    however, since if you knew your way around the Vax source
    code 20 years ago you're good to go with the current OpenBSD
    release. The one thing that pops out is that the
    multiprocessor support isn't all that well instrumented, but
    I figure it's that it's not all that well instrumented -yet-
    and it will be improved in the not-too-distant future.

    The netgraph stuff is extremely cool.
    --
    Melinda Shore - Software longa, hardware brevis - shore@panix.com

    Prouder than ever to be a member of the reality-based community

  2. Re: Is FreeBSD just a sandbox for hackers?

    On Wed, 29 Oct 2008 22:41:57 +0000, Michel Talon wrote:

    > There is no example of a reliable system, either Linux or *BSD, based on
    > source, e.g. Gentoo, FreeBSD, etc. Of course source must be available,
    > but as a dangerous tool, to be used only by hackers.


    I'm not sure what you mean by this. It has always seemed to me that
    there are a lot of FreeBSD users who maintain their systems entirely from
    source. I know that I do. Works great, most of the time. Remember to
    keep your eye on /usr/ports/UPDATING and /usr/src/UPDATING and you'll be
    fine.

    I have a colleague who uses (or used, anyway) Gentoo, so I believe that a
    similar way of working is possible in the Linux world, too.

    Further down-message you mentioned the installer. I think that this is a
    fine example of the fundamental dilemma of open source: most of it is
    written to scratch the itch of someone who is in a position to scratch
    it. Others might have the same itch, but be unable to scratch, and that
    just irritates them, and there's nothing to be done, short of getting
    some funding body (like Google with their summer of code scheme) to pay
    someone to scratch the itch for them. Now, most FreeBSD *users* and
    coders don't spend much (or, indeed any) time installing FreeBSD, so the
    need for a shiny installer is not a significant itch. In fact, those of
    us who have done it a few times feel that the current one's main fault is
    that it attempts too much. Anything beyond partitioning the hard disk
    and installing the base operating system is, IMO, wasted effort.
    Everything else can be done better and more easily from a session logged
    into the newly installed bare system. Installing FreeBSD on a new system
    is far faster and less painful than any other install that I know of...

    Cheers,

    --
    Andrew

  3. Re: Is FreeBSD just a sandbox for hackers?

    Bill Gunshannon wrote:
    > In article <1c4hg4ddcskjg2c6t5la8n1tf5mkfnndbl@4ax.com>,
    > Ranter writes:
    >> On 29 Oct 2008 15:22:16 GMT,
    >> Philip Paeps wrote:
    >>
    >>>Ranter wrote:
    >>>> The build stopped again (hour 47) to let me do more
    >>>> configuration, this time of the Ghostscript drivers. The
    >>>> instructions are that I shouldn't be too stingy, because some
    >>>> other port *might* fail unless it can find a certain driver, and
    >>>> that I should have fun with this new configuration style.
    >>>>
    >>>> For my inconvenience, nearly all the more than 200 (I lost count)
    >>>> entries were pre-checked. I had no idea that other FBSD
    >>>> installations were so well-supplied with printers; I feel
    >>>> positively deprived in only having one for my whole lan.
    >>>>
    >>>> Or perhaps the fun he urges me to have lies in going through and
    >>>> un-checking all those > 200 selections?
    >>>
    >>>If you want handholding instead of a flexible operating system with a friendly
    >>>licensed kernel, perhaps you should take a look at one of the many Linux
    >>>distributions packed with "userfriendly" goo.
    >>>
    >>>I'm afraid you'll have to be prepared to put in a bit more effort if you want
    >>>to use FreeBSD.

    >>
    >> To what combination of factors do YOU attribute Linux's success
    >> compared to FBSD, Philip?

    >
    > I have asked this question of a lot of people. The answer is really
    > simple. Marketing. The people pushing Linux are willing to push
    > and push very hard. The people using (and behind) BSD aren't interested.
    > They are happy to play with it and really don't care if it ever becomes
    > the commercial success that Linux is. Sad really, when you consider that
    > BSD has provavble technical superiority and a much more commercial
    > friendly license than Linux.


    You would be amazed to learn about the commercial success BSD has but which
    you don't hear about because the friendly licence doesn't require it. There
    is BSD code in untold millions of embedded devices being used by untold
    millions of people all the time.

    I don't think anyone in the BSD community is particularly worried about the
    "BSD is dying" priests like "Ranter". I for one am not going to bother
    spending more energy on him. I prefer to try to help people who are prepared
    to read documentation and learn about what they're doing rather than blindly
    typing "make install clean" and hoping it goes well (and waiting 47 hours just
    to complain about it not doing so).

    - Philip

    --
    Philip Paeps Please don't email any replies
    philip@paeps.cx I follow the newsgroup.

    Yesterday is history.
    Tomorrow is a mystery.
    Today is a gift.
    That's why it is called 'present'.

  4. Re: Is FreeBSD just a sandbox for hackers?

    Andrew Reilly wrote:
    >
    > Further down-message you mentioned the installer. I think that this is a
    > fine example of the fundamental dilemma of open source: most of it is
    > written to scratch the itch of someone who is in a position to scratch
    > it. Others might have the same itch, but be unable to scratch, and that
    > just irritates them, and there's nothing to be done, short of getting
    > some funding body (like Google with their summer of code scheme) to pay
    > someone to scratch the itch for them.


    But open source projects are where users really can get what they want.
    It may require getting people together to do it, or learning how to do
    it yourself, or paying someone to do it, or even just entering PRs.

    These all require making an effort, but can benefit lots of users.
    Think of it as the price of FreeBSD.

    --
    Warren Block * Rapid City, South Dakota * USA

  5. Re: Is FreeBSD just a sandbox for hackers?

    On 10/30/2008 04:13 AM, Andrew Reilly wrote:
    > On Wed, 29 Oct 2008 20:12:12 +0000, Ranter wrote:
    >
    >> If this trend continues, FBSD will eventually join Minix in the dustbin
    >> of history. I think that would be a *terrible* thing to happen. And a
    >> pretty stupid one, too, considering I believe that fate to be highly
    >> preventable.

    >
    > You've seen the "BSD is dying" meme on slashdot?
    >
    > BTW, perhaps in some of the 50-ish hours that you've been building GNOME,
    > you might have noticed that both FreeBSD *and* all of the GNOME ports are
    > available, pre-built as packages? Just like those linux distributions
    > that you keep going on about. Sure, lots of BSD folk build from source.
    > I know that I do. But the main reason that I do is so that the source is
    > available for me to debug whatever it is, if I feel like it. I don't
    > imagine that many people feel that particular need.


    IMHO, even if pre-compiled binary packages are available, but updating
    and upgrading a free machine is still a time consuming job compared to
    Fedora, Debian and Arch; all these Linux distributions use their own
    package management systems and none stops you compiling and, or fine
    tuning any package.

    BTW, compiling, analyzing, fine-tuning and, or debugging a package is
    not everyone's cup of tea; why force such absurd policies on to newcomers.

    Why force a full ports collection update, fetchindex and, or portsnap of
    all the 19000+ packages and, or patches when someone is using only
    around 700 of these? The *FreeBSD ports collection* and, or package
    management system in its current incarnation *SUCKS*.

    > Do the owners of the three-wheelers that you hanker to join feel that
    > their vehicles are "dying", even though they are a much smaller group
    > than the owners of, say, Toyotas?


    What apples has to with oranges? What if, someone (or FreeBSD
    developers) does have a sense of feeling and, or does not even know what
    the hell that feeling phenomena is? That one's is not dying, but dead.

    --
    Dr Balwinder S "bsd" Dheeman Registered Linux User: #229709
    Anu'z Linux@HOME (Unix Shoppe) Machines: #168573, 170593, 259192
    Chandigarh, UT, 160062, India Gentoo, Fedora, Debian/FreeBSD/XP
    Home: http://cto.homelinux.net/~bsd/ Visit: http://counter.li.org/

  6. Re: Is FreeBSD just a sandbox for hackers?

    On Thu, 30 Oct 2008 06:40:01 +0530, Balwinder S Dheeman wrote:
    > BTW, compiling, analyzing, fine-tuning and, or debugging a package is
    > not everyone's cup of tea; why force such absurd policies on to
    > newcomers.


    What a strange thing to say. No one is forcing any policy on newcomers.
    There are many ways to solve any problem under Unix. Famous Unix saying:
    "provide mechanism, not policy."

    > Why force a full ports collection update, fetchindex and, or portsnap of
    > all the 19000+ packages and, or patches when someone is using only
    > around 700 of these?


    There is no forcing. The ports collection is not part of the FreeBSD
    base. You're welcome to install applications however you like. The
    ports collection is just one way. Pre-built packages are another way
    that is closely related. There are others.

    You make it sound as though updating the ports collection of 19000+
    packages is more of a chore than some particular 700. That is plainly
    silly: the whole collection is best thought of as a single "database of
    where to find and how to install third-party stuff". Whenever you do an
    update, with csup or cvsup or cvs or rdiff or svn, you only perform one
    operation, and the data transferred over the network is only that part of
    the database that changed since the last time you updated it: usually a
    very small amount of data.

    > The *FreeBSD ports collection* and, or package
    > management system in its current incarnation *SUCKS*.


    You're clearly welcome to your opinion, but I don't share it. I've used
    several package management systems, on several Unix-like platforms. None
    are perfect, all have their strengths and weaknesses. I happen to like
    the tradeoffs that the FreeBSD ports system has made. I particularly
    like the easy visibility of what is going on that comes from storing both
    the database of available packages and the database of installed files as
    simple trees of regular text files.

    Just out of interest, what don't you like about it, besides requiring the
    database (the ports tree) to reside locally on your computer?

    > What apples has to with oranges? What if, someone (or FreeBSD
    > developers) does have a sense of feeling and, or does not even know what
    > the hell that feeling phenomena is? That one's is not dying, but dead.


    I'm afraid you're not making any sense at all here. Could you have a go
    at re-phrasing the question?

    Cheers,

    --
    Andrew

  7. Re: Is FreeBSD just a sandbox for hackers?

    In article ,
    Warren Block writes:
    > Andrew Reilly wrote:
    >>
    >> Further down-message you mentioned the installer. I think that this is a
    >> fine example of the fundamental dilemma of open source: most of it is
    >> written to scratch the itch of someone who is in a position to scratch
    >> it. Others might have the same itch, but be unable to scratch, and that
    >> just irritates them, and there's nothing to be done, short of getting
    >> some funding body (like Google with their summer of code scheme) to pay
    >> someone to scratch the itch for them.

    >
    > But open source projects are where users really can get what they want.
    > It may require getting people together to do it, or learning how to do
    > it yourself, or paying someone to do it, or even just entering PRs.
    >
    > These all require making an effort, but can benefit lots of users.
    > Think of it as the price of FreeBSD.
    >

    The problem is, though, that there are many people "out there"
    who don't want to be bothered with having to think. They want
    to be able to insert a CD into their machine, press the "go"
    button and miraculously have a full-featured operating system
    appear. These same people don't want to be bothered with 3d
    party software installation beyond the minimum effort required
    to have, say, OpenOffice magically appear and work. This is
    what M$ offers and has been moderately successful in delivering
    to the 95% of computer users who can't, don't or won't think
    for themselves.

    That said, I agree that the install process could be somewhat
    better - can't be much easier. ISTR RH of years past offering
    the option of installing a "desktop" or a "server" package, each
    of which consisted of those apps typically found in a "desktop"
    or "server" machine. Clearly, neither option was completely
    satisfactory, but did allow for a fairly straightforward install.

    For myself, I much prefer to install base, then to custom install
    the software _I_ want/need/am curious about. This ability to
    customize _my_ machine to meet _my_ needs is one of the things
    that makes - and keeps - fBSD so attractive - and I've been using
    it since 2.1.7, if that makes any difference.

    Bob Melson

    --
    Robert G. Melson | Rio Grande MicroSolutions | El Paso, Texas
    -----
    Thinking is the hardest work there is, which is the probable
    reason so few engage in it. -- Henry Ford


  8. Re: Is FreeBSD just a sandbox for hackers?

    On 29 Oct 2008 22:43:55 GMT,
    Andrew Reilly wrote:

    >On Wed, 29 Oct 2008 20:12:12 +0000, Ranter wrote:
    >
    >> If this trend continues, FBSD will eventually join Minix in the dustbin
    >> of history. I think that would be a *terrible* thing to happen. And a
    >> pretty stupid one, too, considering I believe that fate to be highly
    >> preventable.

    >
    >You've seen the "BSD is dying" meme on slashdot?


    I don't think so, no. I don't read slashdot except by accident.
    Why, is it there too?

    Where I see it is partly in the *increasing* disparity between
    Linux and BSD in terms of apps and visibility, and partly in what
    I interpret as shakiness in basic support for FBSD - delayed
    releases and, astonishingly, even bad discs from FreeBSDMall. It
    would be nice if I were interpreting the evidence wrong, but I
    don't think I am.

    >
    >BTW, perhaps in some of the 50-ish hours that you've been building GNOME,
    >you might have noticed that both FreeBSD *and* all of the GNOME ports are
    >available, pre-built as packages?


    Yep. That's how I ended up trying to build Gnome - the package
    didn't run.

    Perhaps I should point out, since it's apparently not obvious,
    that the only reason I kept going with that attempted build was
    to see (a) whether it would run to successful completion at all
    and (b) how long it would take, if it did. As I found out, and
    reported here, besides being as Melinda calls it "behemothware",
    the build had two *basic* failings that should never have got
    past the release-engineering group.


    >Just like those linux distributions
    >that you keep going on about. Sure, lots of BSD folk build from source.
    >I know that I do. But the main reason that I do is so that the source is
    >available for me to debug whatever it is, if I feel like it. I don't
    >imagine that many people feel that particular need.
    >
    >Do the owners of the three-wheelers that you hanker to join feel that
    >their vehicles are "dying", even though they are a much smaller group
    >than the owners of, say, Toyotas?


    They'd have to be crazy not to, don't you think? Not that the
    particular car they personally own is dying, but that (especially
    in the US, where 2CVs are thin on the ground) it will always be
    difficult to find parts and that anything they want done to the
    machine they'll have to do personally. It's a hobby for them, a
    way to avoid boredom.

    But a car is a much more limited tool than a computer. It's good
    for carrying things from here to there and that's all. It's not
    a tool for the mind. The owner can't use it to create other
    things or solve other problems. It's useless for any purpose
    other than carrying things, and in the case of an unusual machine
    like the P. some social stroking when showing it off. (Which is
    a good part of the reason why I'll probably never own one: there
    are other ways to carry things, I don't need social stroking, and
    I'm too busy ever to be bored.)

    But maybe your computer is only an anti-boredom hobby for you,
    and you'll continue to play with FBSD even if it becomes a minor
    backwater like Minix?

  9. Re: Is FreeBSD just a sandbox for hackers?

    On Thu, 30 Oct 2008 10:38:55 +0000, Ranter wrote:

    > But maybe your computer is only an anti-boredom hobby for you,


    Well, I won't deny that I enjoy using it, but no, I use the FreeBSD
    systems that I look after for work, at work. They look after themselves
    and "just work". Just ask Yahoo and all of the other big users. Having
    servers that just do what you ask of them is a big advantage.

    --
    Andrew

  10. Re: Is FreeBSD just a sandbox for hackers?

    On Thu, 30 Oct 2008 00:26:08 -0500,
    melsonr@aragorn.rgmhome.net (Robert Melson) wrote:

    >In article ,
    > Warren Block writes:
    >> Andrew Reilly wrote:
    >>>
    >>> Further down-message you mentioned the installer. I think that this is a
    >>> fine example of the fundamental dilemma of open source: most of it is
    >>> written to scratch the itch of someone who is in a position to scratch
    >>> it. Others might have the same itch, but be unable to scratch, and that
    >>> just irritates them, and there's nothing to be done, short of getting
    >>> some funding body (like Google with their summer of code scheme) to pay
    >>> someone to scratch the itch for them.

    >>
    >> But open source projects are where users really can get what they want.
    >> It may require getting people together to do it, or learning how to do
    >> it yourself, or paying someone to do it, or even just entering PRs.
    >>
    >> These all require making an effort, but can benefit lots of users.
    >> Think of it as the price of FreeBSD.
    >>

    >The problem is, though, that there are many people "out there"
    >who don't want to be bothered with having to think.


    This is true.

    But in saying this, I think you're shortchanging another group:
    the ones who don't want to be bothered with having to think
    *about their tools* because everything they've got is already
    going into thinking about an important problem at a higher level
    of aggregation.

    Imagine how it would be if, in order to work on the operating
    system, you had to first design and fab the chips, route and etch
    the pcb, build the drives, etc. And any suggestion for change
    produced a reaction in the range from incomprehension (you mean
    there are people who *don't* want to fab their own chips?) to
    narrow-minded insults. :-)


  11. Re: Is FreeBSD just a sandbox for hackers?

    In article ,
    Ranter wrote:
    >But in saying this, I think you're shortchanging another group:
    >the ones who don't want to be bothered with having to think
    >*about their tools* because everything they've got is already
    >going into thinking about an important problem at a higher level
    >of aggregation.


    "Aggregation?"

    Anyway, I think part of the problem here is that you've
    become dependent on non-standard tools, or rather that you
    can't get work done unless you modify the system a lot. The
    core set of tools has been stable for decades. The add-on
    stuff still isn't stable. I think you're leaving yourself
    awfully vulnerable to bouts of nonproductivity. And you're
    not alone - I know other people who will tell you that
    they're Unix (BSD, Linux, Solaris, whatever) die-hards but
    can't function in a stock Unix environment.
    --
    Melinda Shore - Software longa, hardware brevis - shore@panix.com

    Prouder than ever to be a member of the reality-based community

  12. Re: Is FreeBSD just a sandbox for hackers?

    On 30 Oct 2008 08:00:49 -0400,
    shore@panix.com (Melinda Shore) wrote:

    >In article ,
    >Ranter wrote:
    >>But in saying this, I think you're shortchanging another group:
    >>the ones who don't want to be bothered with having to think
    >>*about their tools* because everything they've got is already
    >>going into thinking about an important problem at a higher level
    >>of aggregation.

    >
    >"Aggregation?"


    That might not be the best word, but I couldn't think of a better
    one. What I meant was an application-level problem, where in a
    logical block diagram of the solution the computer system is
    represented by only one block no matter how many blocks there are
    in toto. What word would you have picked?

    >
    >Anyway, I think part of the problem here is that you've
    >become dependent on non-standard tools, or rather that you
    >can't get work done unless you modify the system a lot. The
    >core set of tools has been stable for decades. The add-on
    >stuff still isn't stable. I think you're leaving yourself
    >awfully vulnerable to bouts of nonproductivity. And you're
    >not alone - I know other people who will tell you that
    >they're Unix (BSD, Linux, Solaris, whatever) die-hards but
    >can't function in a stock Unix environment.


    You might be right, though I think you'd need to say more before
    I'd know whether to fully agree. The last time I really *used*
    fBSD as a tool, I only installed the base system plus (iirc) twm
    because all I needed was the compiler.

    This time I need a tame webserver for the next stage of
    development,so since there isn't an install option that says "AMP
    server with windowed console", I have to grub around installing
    five or six pieces and hope that the people who packaged them
    were better engineers than whoever did up that cursėd Gnome. :-(

  13. Re: Is FreeBSD just a sandbox for hackers?

    Ranter wrote:
    >
    > This time I need a tame webserver for the next stage of
    > development,so since there isn't an install option that says "AMP
    > server with windowed console", I have to grub around installing
    > five or six pieces and hope that the people who packaged them
    > were better engineers than whoever did up that cursėd Gnome. :-(


    Gnome is really the worst example. Some years ago, i wanted to see a
    working Gnome installation on FreeBSD, so i spent a whole week compiling
    it, and i remember very well having to hack into Gnome source code to
    make it compile. Fortunately nowadays there is a FreeBSD Gnome team who
    does this job, so, in principle Gnome compiles, modulo exceptions, and
    binary packages exist. There are very few ports in the 19000 existing
    ports that don't compile, and most of them have binary packages. To make
    an experiment i have waited around 1 year to upgrade the packages on my
    desktop, with around 1000 ports installed. I have upgraded everything
    using precompiled packages from the FreeBSD repositories, using a custm
    script i have. Needless to say, almost everything had to be upgraded,
    and almost everything came in binary form. I had to compile something
    like 10 ports, like pdflib. The whole experiment did not take more than
    a couple of hours. Hence i can safely say that the FreeBSD ports system
    is in good shape. If you want to maintain your installation by
    compiling from source, there is a wonderful tool, portmaster, written by
    Doug Barton. I am using it on some jails with few ports installed and it
    works perfectly. I cannot say the same if you want to use packages, and
    manage them with portupgrade. My opinion is that something like apt-get
    or yum for FreeBSD packages is sorely missing.



    --

    Michel TALON


  14. Re: Is FreeBSD just a sandbox for hackers?

    In article ,
    Ranter wrote:
    >What word would you have picked?


    I think I've had drawn a different diagram. But anyway, if
    all you need is a stable web server, why do you care about
    what platform is underneath it. Are there performance
    issues, scaling issues, ... ?

    >This time I need a tame webserver for the next stage of
    >development,so since there isn't an install option that says "AMP
    >server with windowed console", I have to grub around installing
    >five or six pieces and hope that the people who packaged them
    >were better engineers than whoever did up that cursėd Gnome. :-(


    Do you really need Gnome?
    --
    Melinda Shore - Software longa, hardware brevis - shore@panix.com

    Prouder than ever to be a member of the reality-based community

  15. Re: Is FreeBSD just a sandbox for hackers?

    Ranter wrote:

    >> "Aggregation?"


    > That might not be the best word, but I couldn't think of a better
    > one. What I meant was an application-level problem, where in a
    > logical block diagram of the solution the computer system is
    > represented by only one block no matter how many blocks there are
    > in toto. What word would you have picked?


    The word "abstraction" comes to mind.

    Alphons

    --
    If riding in an airplane is flying, then riding in a boat is swimming.
    If you want to experience the element, get out of the vehicle.

  16. Re: Is FreeBSD just a sandbox for hackers?

    Ranter wrote:
    > But in saying this, I think you're shortchanging another group:
    > the ones who don't want to be bothered with having to think
    > *about their tools* because everything they've got is already


    Flawed logic again.
    Either you select tools you already have learned to use, or you find a
    new tool and learn to use it.
    --
    Torfinn Ingolfsen,
    Norway

  17. Re: Is FreeBSD just a sandbox for hackers?

    On 30 Oct 2008 10:02:22 -0400,
    shore@panix.com (Melinda Shore) wrote:

    >In article ,
    >Ranter wrote:
    >>What word would you have picked?

    >
    >I think I've had drawn a different diagram.


    That's certainly a point of view :-)


    > But anyway, if
    >all you need is a stable web server, why do you care about
    >what platform is underneath it. Are there performance
    >issues, scaling issues, ... ?


    I'm working on a solution for a unique problem with no prior art
    to serve as a guide. The decision has already been taken that it
    will be hosted in a commercial rent-a-NOC for starters, which
    means that the platform will be Windows (eeeuuwww) , Linux or
    FreeBSD. I expect scalability to be the major unavoidable issue,
    with performance, of course, driving it. So, to me, the only
    choice that makes sense is FBSD.

    >
    >>This time I need a tame webserver for the next stage of
    >>development,so since there isn't an install option that says "AMP
    >>server with windowed console", I have to grub around installing
    >>five or six pieces and hope that the people who packaged them
    >>were better engineers than whoever did up that cursėd Gnome. :-(

    >
    >Do you really need Gnome?


    No, not at all. It's just that since I don't positively
    *dislike* upholstered UIs, I thought I'd bring up Gnome, having
    had a non-negative experience with it before, as the console for
    the tame server. But although that worked last time, it didn't
    work this time. Had I not also wanted to try sounding the tocsin
    about the future of FBSD, I'm sure I'd merely have muttered
    "buggrit" and possibly "millenium hand and shrimp" for good
    measure, and brought up the first wm I could put my hand on,
    probably twm (which is approximately what I'm going to do now,
    only it won't be twm since awesome and blackbox have both had
    nice things said about them in this thread).

    FBSD could be in with a good chance against Windows as the o/s of
    choice -- though first it would have to bump Linux out of the
    way. But, surprisingly (surprising to me, maybe not to anyone
    else) it seems that nearly all the vocal energy is on the side of
    letting it continue to fade into obscurity. Weird.

  18. Re: Is FreeBSD just a sandbox for hackers?

    On Thu, 30 Oct 2008 16:01:45 +0000,
    Fonz wrote:

    >The word "abstraction" comes to mind.


    "Abstraction" would have worked. It was the other possibility,
    but at that moment, for some reason, I felt that the concept of
    folding together was more figural than the concept of teasing out
    an essence.

  19. Re: Is FreeBSD just a sandbox for hackers?

    On 2008-10-30, Andrew Reilly was urged to write the following:

    >> The *FreeBSD ports collection* and, or package
    >> management system in its current incarnation *SUCKS*.

    >
    > You're clearly welcome to your opinion, but I don't share it. I've used
    > several package management systems, on several Unix-like platforms. None
    > are perfect, all have their strengths and weaknesses. I happen to like
    > the tradeoffs that the FreeBSD ports system has made. I particularly
    > like the easy visibility of what is going on that comes from storing both
    > the database of available packages and the database of installed files as
    > simple trees of regular text files.


    Totally seconded. I'm a Debian user, but whenever I find the time I
    tinker with FreeBSD. Over time I grew very fond of the way the ports
    collection works. The control you've got over how the software will be
    installed is great. I'm even considering a migration to Slackware, as
    I understand they have a similar approach to installing software.

    --
    Religion is what keeps the poor from murdering the rich.
    ~ Napoleon Bonaparte

  20. Re: Is FreeBSD just a sandbox for hackers?

    On Thu, 30 Oct 2008 12:48:04 GMT, Ranter wrote:
    > You might be right, though I think you'd need to say more before I'd know
    > whether to fully agree. The last time I really *used* fBSD as a tool, I
    > only installed the base system plus (iirc) twm because all I needed was
    > the compiler.
    >
    > This time I need a tame webserver for the next stage of development,so
    > since there isn't an install option that says "AMP server with windowed
    > console", I have to grub around installing five or six pieces and hope
    > that the people who packaged them were better engineers than whoever did
    > up that cursed Gnome. :-(


    That's, unfortunately, one of the characteristics of not making choices for
    you, but empowering *you* to make the choices for yourself. This is the
    traditional ``provide tools not policies'' philosophy of FreeBSD:

    Instead of bundling a `standard httpd' into the base system, FreeBSD gives
    you a stable `base' system with a well-known and very slowly changing set
    of really important tools, and then the Ports. This means that you have
    extra work to do to:

    * Pick a web server implementation

    * Install it from the pre-built Packages or build it from Ports

    * Configure it to match your particular needs

    A case in point is that when I recently wanted to experiment with the Wiki
    software that runs EmacsWiki[1], I had to install lighttpd myself,
    experiment with its configuration until I got the request rewrite rules
    right, and then dump a bunch of Perl CGI scripts in its runtime area.

    This sounds like too much work, but I was happy that I could install just
    the bits that I wanted and *nothing* more. If all the choices were
    pre-made for me, I wouldn't like the result a lot, because I usually like
    knowing exactly *what* I installed. It is this ability that makes FreeBSD
    so attractive for me.

    Having said that, I understand that there are people who do not have either
    the time or the inclination to think about all the choices involved in
    installing an HTTP server. That's fine too, and it is well known that
    sometimes too many choices are a major hindrance of progress by the sheer
    amount of effort needed to weed through all the options[2].

    If *that* is what you are hinting at by ``I have to grub around installing
    five or six pieces'', then you do have a very valid point. Unfortunately,
    I don't know if there is a good way of keeping the ability to make those
    choices *and* make things easier for the crowds who want to pop in a CD-ROM
    select a check-box or two, hit `Next' a couple of times and then go have a
    coffee, tea or something while their new web server is installed :-(

    References
    ----------

    [1] http://www.emacswiki.org/emacs/
    [2] Barry Schwartz. ``The Paradox of Choice.''
    http://video.google.com/videoplay?do...89424111915182

+ Reply to Thread
Page 2 of 3 FirstFirst 1 2 3 LastLast