Lyx sux. It's not wysiwyg, it's slow, it has no built-in extentions (rudimentary CAD, flow-charting, etc.), few import abilities and it controls the way your output looks too much for an office user, due to it's typesetting nature. If you need to write a science article - that's just what the doctor ordered, but if you want to write a document with lot's of little drawings which you want to create inside the wordprocessor itself then you are out of luck. And pagination is a big pain - TeX does not do a good job of pagination unless you manually rearrange things and for this you need to compile things and pipe them to ghostview: this is slow and requires an external viewer, which is an extra layer of complexity. Office suites are meant at people who are uncofortable with using a mouse because it represents interacting with this complicated thing called computer. Else they'd use vi, which is better than Lyx at TeX (or any) editing anyway.
Ideological arguments aside, the big deal is that their pricing is only dictated by market conditions. Thus, if they beat their competition (Microsoft and Corel), they could start charging Microsoft scale money for it. If you are looking for products that are "cheap/free" then nothing but OS/FS will do in the long run. However I tend to agree that ideological arguments mostly fail for office suites: the people who use them would usually not obtain any empowerment or freedom by having access to code. For a secretary, the spell-checker either works or it doesn't - s/he will not even imagine fixing it. In general I feel that software aimed at people who are not aware of why or how it runs, who have not heard of source code and who emphatically do not want to get into this stuff (examples would be office suites, games, organizers, banking software) - such software need not be open-source ideologically, but it does need to be open source for pricing reasons. If most users are unlikely to scratch their own itch, then open source is equivalent to or even worse than closed source, in the sense that such open source software will be underdeveloped (due to lack of programmers' interest) compared to its closed brethern.
>>I hope for your sake I don't find out where you live! Just because there are some weirdos here, does not mean you (or anyone for that matter) should go stoop to their level (threats and all). Besides, these people need a psychiatrist, not a chiropractor.
As you said, for central limit theorem you need a lot of independent rv's. I wonder if self-similarity causes interdependent rv's, such that they can be shown to converge to a specific curve, or if their research is just: "look at our data - it looks about the same" type stuff. Does anyone know if there is a formal theoretical basis for this work?
1. As far as I understand, all window decorations are up to X/window manager, whereas with MDI they would be up to the app itself - a big difference. You keep thinking of MDI as having many windows inside one big one - think the way Java allows you - sub windows can really be subclassed from a panel class and be directly controlled by the app, not X. 2. Without doing a lot of custom setting up, how do I - oh say - auto cascade all my Netscape windows, or auto tile all my xterms? I do not want Netscapes tiled, just xterms, and vice versa. 3. see 1 4. Duh
Your comments like "what could be simpler" a few posts above, or flaming people as "idiots" here, show that you are a snob and a rather intolerant at that. Do you hope to attract people with flames like that?
I do not have a Windows partition nowadays, but one thing I miss from Windows is MDI. It has many advantages: 1. You can have special window decorations for child windows. 2. You can auto-arrange relevant windows in a variety of ways. 3. This should be faster, at least as far as Z-ordering goes. 4. If you ever used framebuffer with split view, you love MDI and don't even know it.
I work at Beckman as a physics grad student. Our lab is on the other side of the wall from the "cave", i.e. immersive 3-d display (VR) system. As far as I can tell all the power is used to play games, at least that's the only sounds that come from there. OTOH, the number of SGI's per square inch is impressive.
>>Gui dialog designer that produced NO code. We stored dialog >>layouts as data that an object factory would then instantiate.
So kind of like an XML file then?
>>A functionality-to-widget binding method that did not require >>the code to find a particular dialog item. You just needed to know >>the name of the widget and you bound your functionality to its name.
Berlin guys were discussing the "tasket" concept which seems to be what you are describing here. You may have valueable knowledge for that project.
Well, I shut down my Linux box regularly. I live in a small apartment, so my computer is next to my bed and I can't sleep with fans being as noisy as they are (OK, so the cover on my case is permanently off:).
First bear in mind that both KDE and CNOME are work in progress. Gnome does not deserve to be beyond version 0.1, while KDE just made it to 1.0 with 1.1.1 release. Neither can be taken as a good example of Linux desktop. KDE 2.0 promises some real apps, and so does GNOME 2.0. They will probably have enough features to be competitive by version 3.0, by which time they may run same CORBA backend and same dnd so coding for one would be roughly the same as for another (esp. if KDE adds more language bindings, regardless of how many people need it). Most people in Linux world do use zip (gzip), so I am not sure what the difference is, except that winzip is not available (bfd). Most compressed programs you'll see have extention.gz, so there is quite a bit of uniformity there. As far as APIs, it is not clear that it is a good thing to have only one. Besides, they are in no way a part of the "end-user world". I do think LSB is good, and it would be better if it were folded into posix, so that noone out there could ignore it. But one set of widgets? Yuck. If people listened to you, we'd be using Motif without any alternatives. IMHO, that's worse than all Windows crashes times 100.
Quote: "given a large number N, if you know that N has two factors, you can find them by trying to divide N by every number up to N/2". Apparently no math major has reviewed their work (should be "up to [sqrt(N)]", where [] denotes integer floor function).
The biggest concern is not just money, but in general that the motivating factors like money or fame tend to sway development in a particular direction. Why is USB still not standard in Linux, why is scanner support virtually non-existant, why is sound support highly limited? All because those are the fundamentals that don't bring fame or fortune. It's the dirty work everyone hates. Let me also mention winmodems which are still not supported even though many models have been around for months if not years. Mozilla development is lagging behind because it was not sexy enough to attract a significant number of outside developers. I mean I just keep coming up with more examples of what current OSS or FS developers are "not motivated" to code. Suck stumbled upon a real problem even though it has little to do with Redhat IPO.
Your last sentence made me curious: is there a SCSI card that plugs into an AGP port? Most SCSI devices need high bandwidth and often have memory requirements, else you could use a cheaper PCI. So how about an AGP to UW-SCSI adapter? Anyone?
Re:I think they are going in the wrong direction h
on
The Future of KDE
·
· Score: 1
Thanks for the tip. I'll try since I am forced to use a Mac in the lab. But this calls into question how intuitive Mac UI really is, given that I couldn't find a way to do it despite looking hard.
Slashdot Mirror
Ok, I got no clue about the subject, which is why I am wondering:
does TWAIN plugin mean SANE is now not needed?
Lyx sux. It's not wysiwyg, it's slow, it has no built-in extentions (rudimentary CAD, flow-charting, etc.), few import abilities and it controls the way your output looks too much for an office user, due to it's typesetting nature. If you need to write a science article - that's just what the doctor ordered, but if you want to write a document with lot's of little drawings which you want to create inside the wordprocessor itself then you are out of luck. And pagination is a big pain - TeX does not do a good job of pagination unless you manually rearrange things and for this you need to compile things and pipe them to ghostview: this is slow and requires an external viewer, which is an extra layer of complexity. Office suites are meant at people who are uncofortable with using a mouse because it represents interacting with this complicated thing called computer. Else they'd use vi, which is better than Lyx at TeX (or any) editing anyway.
Ideological arguments aside, the big deal is that their pricing is
only dictated by market conditions. Thus, if they beat their
competition (Microsoft and Corel), they could start charging
Microsoft scale money for it. If you are looking for products that
are "cheap/free" then nothing but OS/FS will do in the long run.
However I tend to agree that ideological arguments mostly fail for
office suites: the people who use them would usually not obtain
any empowerment or freedom by having access to code. For a
secretary, the spell-checker either works or it doesn't - s/he will
not even imagine fixing it.
In general I feel that software aimed at people who are not aware of
why or how it runs, who have not heard of source code and who
emphatically do not want to get into this stuff (examples would be
office suites, games, organizers, banking software) - such software
need not be open-source ideologically, but it does need to be
open source for pricing reasons. If most users are unlikely to scratch
their own itch, then open source is equivalent to or even worse than
closed source, in the sense that such open source software will be
underdeveloped (due to lack of programmers' interest) compared to
its closed brethern.
>>I hope for your sake I don't find out where you live! Just because there are some weirdos here, does not mean you (or anyone for that matter) should go stoop to their level (threats and all). Besides, these people need a psychiatrist, not a chiropractor.
You can't trust another party, big or small.
As you said, for central limit theorem you need a lot of
independent rv's. I wonder if self-similarity causes
interdependent rv's, such that they can be shown to
converge to a specific curve, or if their research is just:
"look at our data - it looks about the same" type stuff.
Does anyone know if there is a formal theoretical
basis for this work?
It's called Teledesic and Microsoft owns it.
How about full screen anti-aliasing? If it's in, this
card will be hard to beat. If it's out, I'll pass.
Ok, I am sorry, I did not mean to say that
/. is misbehaving. This may be a repeat
he actually was a snob, just that he made a
very strong impression to that effect in his
posts.
P.S.
post.
1. As far as I understand, all window
decorations are up to X/window manager,
whereas with MDI they would be up to
the app itself - a big difference.
You keep thinking of MDI as having many
windows inside one big one - think the way
Java allows you - sub windows can really
be subclassed from a panel class and be
directly controlled by the app, not X.
2. Without doing a lot of custom setting up,
how do I - oh say - auto cascade all my
Netscape windows, or auto tile all my xterms?
I do not want Netscapes tiled, just xterms,
and vice versa.
3. see 1
4. Duh
Your comments like "what could be simpler"
a few posts above, or flaming people as
"idiots" here, show that you are a snob
and a rather intolerant at that. Do you
hope to attract people with flames like
that?
I do not have a Windows partition nowadays,
but one thing I miss from Windows is MDI.
It has many advantages:
1. You can have special window decorations
for child windows.
2. You can auto-arrange relevant windows in
a variety of ways.
3. This should be faster, at least as far
as Z-ordering goes.
4. If you ever used framebuffer with
split view, you love MDI and don't even
know it.
My understanding was that it was Sun's
architecture for slapping a lot of
processor cores on the same silicon.
Does this mean that Sun's MAJC architecture (with say
1024 processors on one chip), would also be ideal?
I work at Beckman as a physics grad
student. Our lab is on the other side
of the wall from the "cave", i.e. immersive
3-d display (VR) system. As far as I can tell
all the power is used to play games, at least
that's the only sounds that come from there.
OTOH, the number of SGI's per square inch
is impressive.
Seems like a security threat to me.
>>Gui dialog designer that produced NO code. We stored dialog
>>layouts as data that an object factory would then instantiate.
So kind of like an XML file then?
>>A functionality-to-widget binding method that did not require
>>the code to find a particular dialog item. You just needed to know
>>the name of the widget and you bound your functionality to its name.
Berlin guys were discussing the "tasket" concept which seems to be
what you are describing here. You may have valueable knowledge
for that project.
Well, I shut down my Linux box regularly. I live in a small :).
apartment, so my computer is next to my bed and I can't sleep
with fans being as noisy as they are (OK, so the cover on my
case is permanently off
First bear in mind that both KDE and CNOME are work in progress. .gz, so there is
Gnome does not deserve to be beyond version 0.1, while KDE just
made it to 1.0 with 1.1.1 release. Neither can be taken as a good
example of Linux desktop. KDE 2.0 promises some real apps, and
so does GNOME 2.0. They will probably have enough features to
be competitive by version 3.0, by which time they may run same
CORBA backend and same dnd so coding for one would be roughly
the same as for another (esp. if KDE adds more language bindings,
regardless of how many people need it).
Most people in Linux world do use zip (gzip), so I am not sure what
the difference is, except that winzip is not available (bfd).
Most compressed programs you'll see have extention
quite a bit of uniformity there.
As far as APIs, it is not clear that it is a good thing to have only one.
Besides, they are in no way a part of the "end-user world".
I do think LSB is good, and it would be better if it were folded into
posix, so that noone out there could ignore it. But one set of widgets?
Yuck. If people listened to you, we'd be using Motif without any
alternatives. IMHO, that's worse than all Windows crashes times 100.
Quote:
"given a large number N, if you know that N has two factors, you can
find them by trying to divide N by every number up to N/2".
Apparently no math major has reviewed their work (should be "up to [sqrt(N)]",
where [] denotes integer floor function).
I suppose wearable computing would be a better answer, since you'd never need
to part with any data or hardware at any time.
The biggest concern is not just money, but in general that the motivating
factors like money or fame tend to sway development in a particular
direction. Why is USB still not standard in Linux, why is scanner support
virtually non-existant, why is sound support highly limited? All because
those are the fundamentals that don't bring fame or fortune. It's the dirty work
everyone hates. Let me also mention winmodems which are still not supported
even though many models have been around for months if not years.
Mozilla development is lagging behind because it was not sexy enough to attract
a significant number of outside developers. I mean I just keep coming up with
more examples of what current OSS or FS developers are "not motivated" to code.
Suck stumbled upon a real problem even though it has little to do with Redhat IPO.
So it looks like Woz was wrong. There were computers with integrated
keyboard (IBM 5100) before Apple I.
Your last sentence made me curious: is there a SCSI card that plugs
into an AGP port? Most SCSI devices need high bandwidth and often
have memory requirements, else you could use a cheaper PCI. So
how about an AGP to UW-SCSI adapter? Anyone?
Thanks for the tip. I'll try since I am forced to use a Mac in the lab.
But this calls into question how intuitive Mac UI really is, given that
I couldn't find a way to do it despite looking hard.