Proprietary vs Open source, Episode MVXXI
on
StarOffice 6.0
·
· Score: 1
And if you don't care about Microsoft having any competition in office suite software, then by all means download
OpenOffice in lieu of buying StarOffice.
So you are suggesting we make a donation to
Sun Microsystems? I already did that, I bought
some of their %^%## stock.
But seriously, if Microsoft products really are
such scheise (could be, don't use em myself) then
a competitor with a good product should not
have to really on idealogically motivated sympathy.
The original "open source business model" was that
we could have it both ways: someone to support
us, for $, and the source, to modify, or
compile for whatever weird ass platform we happen
to be interested in.
Whether this is a good business model will be
decided by the holy markets, but it certainly
is attractive to a savvy consumer.
Personally the main reason I don't like proprietary software is that I find it to be
an adminstrative pain in the ass.
License managers?
Purchase orders?
Support on non-intel platforms?
Dependance on red hat 6.2? or windows '95
versus, sit down at a computer, install the tools
I need, and start working.
Your mileage may vary, and probably does.
For those of you who skipped to the end looking for "the point", here it is
There are non-monetary, and non-ideological reasons people might
prefer Open Office to StarOffice.
Devotees of Adam Smith can't have it both ways. Either Sun offers something worth $79 or not. Ask yourself, would you buy this product
from Microsoft?
Commies like Walter Hewlett. Commies
like the (many) shareholders who dumped their stock
in both companies. Commies like the many who
wonder what the heck happened to Compaq+DEC.
PDF is swell as a format to print from. Or squint
at on the screen, but it doesn't exactly edit
nicely.
I don't know what the answer is for editable
collaboration between word-users and the TeX-users. Part of the problem is that all
of the conversions are (as you note) rather lossy.
People who use TeX tend to care.
People who use Word or WP for documents with
any math in them by definition don't care what
they look like:-)
Depending on the politics of the situation, it
may be feasible for the one person to use
LyX
(or perhaps TeXmacs) and the other to edit TeX.
This gets over (more or less) the "TeX is hard" objection, but it does require a pretty big hassle for Windows users (Install an X server? Is that
for Pr0n?).
Ummm. Well, 16MB is not that low, by the
standards of the time. I ran Sun Sparcstations
(ELC) and the roughly equivalent 486 DX2/66
with this amount of memory.
They both ran X fine. KDE or GNOME, no. X yes.
So maybe you should petition to opensource
Linux 0.whatever or FreeBSD 2.2.5:-)
SunOS 4.1 was also pretty lean. Obligatory
flamewar about slowlaris omitted due to extreme boredom.
Well this, and the article it replies to,
pretty well sum it up. Either C is the best
tool for the job, or not. If you don't care
about performance, then probably not.
I don't mean this to come across as a put down,
but there are _lots_ of applications that
are nothing like the three IO-bound ones that
you mention. And although the mass market is
primarily interested in games and office applications, there are lots of people in the
world who still use computers to compute things.
For the last month I have had $100,000 worth of
computers crunching away on a problem in Discrete
Geometry. This means that the basic operation in
my program has been repeated about 150 billion times. So yeah, I care how fast that operation
is (50 machine cycles, last time I checked).
Re:Transition (FreeBSD vs. Debian wrt packages)
on
Debian NetBSD
·
· Score: 1
I too run FreeBSD and debian/linux.
I would say package/port breadth is a tossup.
But I really like the separation of the upgrade
process into core and ports that FreeBSD has.
I have never made my machine unusable using ports. I have had some close calls with apt.
Also, my limited experience is that it is easier
to wedge things with apt-get than with ports
because problems show up during the build process
with ports, and the install process with
apt.
You may cry that that is an unfair comparision
since I could build packages from source on debian. But that is not the typical user experience.
2 things that people have mentioned in choosing
DUnix over Linux.
1) Compilers
The compaq compilers are available under
linux.
2) SMP scalability.
This possibly is a bit of a red herring, since the SMPs are comparatively small (4 processors).
Another issue that I didn't see mentioned is
page coloring, and the VM system in general (hi Greg!). Many people find that there is a 10-20%
improvement on numerical codes.
Another issue is the familiarity of the Compaq
service organization with Tru64 vs. Linux.
Since it doesn't really cost them anything
to include the licenses, and it is probably
good publicity for Compaq's big iron sales
(remember, this was before Compaq decided to
flush all of its HPTC customer down the toilet)
why not?
Do send applicable changes/patches to the original author/maintainer for inclusion in next release of the code. This will only make your job that much easier for the next release.
As I have to explain over and over to my
students and OOO (over object oriented) colleagues:
Computers are about algorithms, get over it.
Hmm. Maybe I am agreeing with you in some sense.
To bring this back on topic, my own experience
is that starting with Java means a whole load
of concepts (inheritence etc...) that didn't
used to be there. This may be the right thing
in the long run, but in first year it means peoples comprehension of loops, branches and arrays is weak. Without that stuff, they can't
really do anything interesting.
It seems clear to me that M$ is going to push hard to prevent university staff from using or contributing to the GPL code base.
I think (hope) this is unlikely to work. The
reason that I personally am an academic is that
I don't have to put up with very much of that
sort of BS. This partly because the tradition
of academic freedom, and partly because universities just don't exercise that kind of
centralized control over their staff; the structures just don't exist.
Now if the government gets involved, wierd shit
can happen. But it won't happen quietly.
Umm. OK. So Microsoft is good because they
made a lot of people rich. That does not
mean their software is not crap. Nor (as
many people, including the odd judge, have
observed) that their business practices
are fair.
I'm going to ignore anything talking about
communism. That isn't
even much of an insult, nevermind an argument.
IBM micro-drives work in an iPaq, under Linux
or otherwise. The battery life is not so
hot as I understand it. If you are really
interested read the mailing list archives
on handhelds.org
Re:I nearly got into a FISTFIGHT with the Amaya fo
on
W3 Releases Amaya 4.0
·
· Score: 1
Heh, and I guess we just got a sample of your
great HTML skills above. Oh. You meant to
have the rest of your article in italics. No
wonder perfect strangers punch you
The articles I put on
my various Web sites are not intended to help people who just want to live a quiet comfortable life (I'm not an expert on this). They are intended to
help young people turn into Linus Torvalds or Richard Stallman or Dan Bricklin and Bob Frankston (Visicalc).
Heh. What he said. You want fun? Try
telling university computer science
professors what software
to run on their computers. Some of us have
a slashdot sized chip on our shoulder. Especially
if the directive comes from computing services.
Light fuse here...
But yeah, I think the poster that started
this chain must be suffering from terminological
inexactitude.
The fact is that open source needs to focus more on the customer side of things if it is ever
going to get corporate and public acceptance, and the $$$ that these things bring with
them.
If you have customers, listen to them, or
make them buy what you already have.
If you don't have customers, then the question
is kind of silly isn't it?
Doesn't this seem pretty much independant
of Open Source/Free Software/Proprietary/
Whatever?
In fact, the open source "pay for enhancements"
model of software business could hardly be
more customer focussed. Rather than some
imaginary customer dreamed up by the marketers,
here you have a real customer, willing to pay
real money, for specific features.
Does this model work for Joe Sixpack?
Probably not. Should we care?
Try it out. The compilers and machines to run
them on, are available for free
(as in beer). OK, so you can't keep the
machine.
For several of my (integer) applications,
an alpha 21264 at 466 Mhz is literally twice as fast
as a pentium III at 500 MHz speed.
Something like 20% of this is in the compilers.
For one application, the fact that the alpha is
64 bit is the killer point. For the other, I'm
not sure what the issue is. The tests are on
the smallest cache size 21264 currently shipping,
which is 2M.
As for what is the best price/performance ratio,
this is a tougher call. Remember that to most
people one system twice as fast is worth lots more
than 2 systems once as fast. And yes, especially
in a beowulf.
One area where alpha _systems_ really clean up is
in memory bandwith and (64bit) PCI bandwidth.
For applications like cplant where the network
(myrinet) is faster than a PC PCI bus, this is a killer factor.
The method described here, called "Standards",
allows computer users to interchange data
freely without being restricted to using software
from a particular manufacturer. In particular
we show how a standard markup langage
(OPML == Our patented markup language) allows the
viewing of web pages by arbitrary browsers,
including those not made by Microsoft.
So use active-X or we will sue your ass!
As a confirmed alpha bigot, I can confirm
that I was pleasantly surprised by the IBM SP
nodes. BUT, only the performance on your code
matters. I still think for me that an alpha
linux cluster is the most cost effective way
to go; with the SP you are really paying for
1) the high speed interconnect
2) The fact that the software works nicely together.
Other than that, it is just another cluster.
If you haven't yet figured out the difference
between clusters, uniform access shared memory
machines, and non-uniform shared memory machines
I recommend reading the first few chapters of
"In search of clusters" by Pfister
Out of curiousity, were you considering moving from Seattle? I have lived in both; I think the differences are not so huge (except taxes!).
Very high cost of living. Notably including the cost of real estate.
What boggles my mind about seattle is the cost of living in the east side 'burbs. Chacun son gout.
Heavy traffic. Decent locations to work tend to be in the downtown, and access is controlled by about half a dozen bridges.
The situation in Seattle is symmetric, but worse. The nice places to live (IMHO) are downtown, and 99.9% of the jobs are on the eastside, with access via two bridges. Well, the traffic in vancouver really is worse though. But the Japanese food is better;-)
Slashdot Mirror
So you are suggesting we make a donation to Sun Microsystems? I already did that, I bought some of their %^%## stock.
But seriously, if Microsoft products really are such scheise (could be, don't use em myself) then a competitor with a good product should not have to really on idealogically motivated sympathy.
The original "open source business model" was that we could have it both ways: someone to support us, for $, and the source, to modify, or compile for whatever weird ass platform we happen to be interested in.
Whether this is a good business model will be decided by the holy markets, but it certainly is attractive to a savvy consumer. Personally the main reason I don't like proprietary software is that I find it to be an adminstrative pain in the ass.
- License managers?
- Purchase orders?
- Support on non-intel platforms?
- Dependance on red hat 6.2? or windows '95
versus, sit down at a computer, install the tools I need, and start working.Your mileage may vary, and probably does. For those of you who skipped to the end looking for "the point", here it is
Gotta run. Have to go rally the proleteriat.
PDF is swell as a format to print from. Or squint at on the screen, but it doesn't exactly edit nicely. I don't know what the answer is for editable collaboration between word-users and the TeX-users. Part of the problem is that all of the conversions are (as you note) rather lossy. People who use TeX tend to care. People who use Word or WP for documents with any math in them by definition don't care what they look like :-)
Depending on the politics of the situation, it
may be feasible for the one person to use
LyX
(or perhaps TeXmacs) and the other to edit TeX.
This gets over (more or less) the "TeX is hard" objection, but it does require a pretty big hassle for Windows users (Install an X server? Is that
for Pr0n?).
Does anyone know of an out-of-the box quiet alternative? I am intersted in a compact
compute server, so graphics and sound are not really a biggie.
Ummm. Well, 16MB is not that low, by the
:-)
standards of the time. I ran Sun Sparcstations
(ELC) and the roughly equivalent 486 DX2/66
with this amount of memory.
They both ran X fine. KDE or GNOME, no. X yes.
So maybe you should petition to opensource
Linux 0.whatever or FreeBSD 2.2.5
SunOS 4.1 was also pretty lean. Obligatory
flamewar about slowlaris omitted due to extreme boredom.
Well this, and the article it replies to, pretty well sum it up. Either C is the best tool for the job, or not. If you don't care about performance, then probably not. I don't mean this to come across as a put down, but there are _lots_ of applications that are nothing like the three IO-bound ones that you mention. And although the mass market is primarily interested in games and office applications, there are lots of people in the world who still use computers to compute things. For the last month I have had $100,000 worth of computers crunching away on a problem in Discrete Geometry. This means that the basic operation in my program has been repeated about 150 billion times. So yeah, I care how fast that operation is (50 machine cycles, last time I checked).
I too run FreeBSD and debian/linux.
I would say package/port breadth is a tossup.
But I really like the separation of the upgrade
process into core and ports that FreeBSD has.
I have never made my machine unusable using ports. I have had some close calls with apt.
Also, my limited experience is that it is easier
to wedge things with apt-get than with ports
because problems show up during the build process
with ports, and the install process with
apt.
You may cry that that is an unfair comparision
since I could build packages from source on debian. But that is not the typical user experience.
2 things that people have mentioned in choosing
DUnix over Linux.
1) Compilers
The compaq compilers are available under
linux.
2) SMP scalability.
This possibly is a bit of a red herring, since the SMPs are comparatively small (4 processors).
Another issue that I didn't see mentioned is
page coloring, and the VM system in general (hi Greg!). Many people find that there is a 10-20%
improvement on numerical codes.
Another issue is the familiarity of the Compaq
service organization with Tru64 vs. Linux.
Since it doesn't really cost them anything
to include the licenses, and it is probably
good publicity for Compaq's big iron sales
(remember, this was before Compaq decided to
flush all of its HPTC customer down the toilet)
why not?
dmalloc (www.dmalloc.com) seems to work pretty well for finding memory leaks. It is distributed under a BSDish
license.
Compiles and runs out of the box on an alpha
running Linux.
GUI? uh no. It has a nifty command line utility to control logging etc...
Hmm. Maybe I am agreeing with you in some sense.
To bring this back on topic, my own experience is that starting with Java means a whole load of concepts (inheritence etc...) that didn't used to be there. This may be the right thing in the long run, but in first year it means peoples comprehension of loops, branches and arrays is weak. Without that stuff, they can't really do anything interesting.
Umm. OK. So Microsoft is good because they made a lot of people rich. That does not mean their software is not crap. Nor (as many people, including the odd judge, have observed) that their business practices are fair. I'm going to ignore anything talking about communism. That isn't even much of an insult, nevermind an argument.
IBM micro-drives work in an iPaq, under Linux or otherwise. The battery life is not so hot as I understand it. If you are really interested read the mailing list archives on handhelds.org
Yep, I've heard of them. Who are you again?
Heh. What he said. You want fun? Try telling university computer science professors what software to run on their computers. Some of us have a slashdot sized chip on our shoulder. Especially if the directive comes from computing services.
Light fuse here...
But yeah, I think the poster that started this chain must be suffering from terminological inexactitude.
Right, or you could just move to Canada :-)
At least until the black helicopters
make us enact the new world order legislation.
Quick, leave while it still legal to take
bathroom breaks during the commercials.
If you don't have customers, then the question is kind of silly isn't it?
Doesn't this seem pretty much independant of Open Source/Free Software/Proprietary/ Whatever?
In fact, the open source "pay for enhancements" model of software business could hardly be more customer focussed. Rather than some imaginary customer dreamed up by the marketers, here you have a real customer, willing to pay real money, for specific features.
Does this model work for Joe Sixpack? Probably not. Should we care?
For several of my (integer) applications, an alpha 21264 at 466 Mhz is literally twice as fast as a pentium III at 500 MHz speed. Something like 20% of this is in the compilers. For one application, the fact that the alpha is 64 bit is the killer point. For the other, I'm not sure what the issue is. The tests are on the smallest cache size 21264 currently shipping, which is 2M.
As for what is the best price/performance ratio, this is a tougher call. Remember that to most people one system twice as fast is worth lots more than 2 systems once as fast. And yes, especially in a beowulf.
One area where alpha _systems_ really clean up is in memory bandwith and (64bit) PCI bandwidth. For applications like cplant where the network (myrinet) is faster than a PC PCI bus, this is a killer factor.
The method described here, called "Standards", allows computer users to interchange data freely without being restricted to using software from a particular manufacturer. In particular we show how a standard markup langage (OPML == Our patented markup language) allows the viewing of web pages by arbitrary browsers, including those not made by Microsoft. So use active-X or we will sue your ass!
As a confirmed alpha bigot, I can confirm that I was pleasantly surprised by the IBM SP nodes. BUT, only the performance on your code matters. I still think for me that an alpha linux cluster is the most cost effective way to go; with the SP you are really paying for 1) the high speed interconnect 2) The fact that the software works nicely together. Other than that, it is just another cluster. If you haven't yet figured out the difference between clusters, uniform access shared memory machines, and non-uniform shared memory machines I recommend reading the first few chapters of "In search of clusters" by Pfister
One thing nmap does not seem to check for
is TCP wrappers; it reports a port open,
but TCP wrappers may drop all connects to that
port.
Just to clarify: tcsh has nothing to do with
GNU, the GPL, or the FSF as far as I can tell
from looking at the source (on my FreeBSD 4.1-RC
system)
less(1) on the other hand seems to be
GPLed and copyright the FSF.