It's an interesting viewpoint, but I think you're placing too much emphasis on the physical aspects of a library. A library is just as much any collection of books or documents under communal ownership (your taxes make you a partial owner of your town's collection of books).
For a second, let's make the arguably unpleasant assumption that an artificial scarcity will be maintained amongst digital documents in the future (as foreshadowed by the current Ebook scandal). In such a situation, it would still be possible for groups of people to get together online and make their own private document sharing collective, in which the right for any one person to read a certain book at a certain time was passed around like a token. This would be functionally the same as a library today, but in the pure digital world.
This is obviously very different from the utopic
vision of unlimited copies of digital documents being freely available to all. I don't mean to pass judgement one way or the other (since my opinion doesn't count for much anyway), but until such time as the world figures out a practical incentive for authors, composers, artists, etc to keep producing, then a library like this would not be such a bad idea.
Some search engines have this capability built in, including the one which I work on professionally: AltaVista Enterprise Search.
But my grandest creation, as history will tell,
Re:Avoid PDF for On-Screen Reading
on
PDF Alternatives?
·
· Score: 1
While Nielsen is fairly widely respected, this particular editorial of his is an absolute embarassment.
His criticisms fall into two categories:
Problems with Adobe Acrobat, the most common PDF viewer
By default, it uses a paged rather than a continuous scrolling model.
It does not support incremental rendering while downloading.
The Macintosh port is somewhat crash-prone.
Problems with what people put in their PDF documents, especially, in his opinion, too few internal navigational links.
None of these, of course, is a deficiency of the file format itself, and all are easily addressed. I'm not an apologist of any sort, but misrepresentations like this are not particularly useful.
Speaking from experience, I can say that posting HTML form data (using either GET or POST) works just fine in arbitrary encodings. The encoding will always be that of the page containing the form.
If your script is the one capturing the form data, then it is also usually the one which generated the page with the form on it, so you can tell the browser to switch into whatever encoding you want (using the charset option on the Content-type HTTP header or placing it in an HTML META tag). But my grandest creation, as history will tell,
Wrong. You're confusing the UCS-2, the encoding, with UCS, the character repertoire.
To spell it out, "Universal Character Set - Two Byte Encoding" (UCS-2) is one of many encodings which can represent the "Basic Multilingual Plane" (BMP) subset of the "Universal Character Set" (UCS) character repertoire.
If by 2030 we have invented a matter duplicator that's as cheap as copying a CD today, will we outlaw it and drive it underground?
There was an old science fiction book by the name of A for Anything which explores the societal implications of just such a technology. It does not take anywhere's near as cheery a view as the EFF.
I haven't decided exactly where I stand on the issue, but Mr. Gilmore raises some good points.
Div.
I would much sooner moderate this as "informative" than "off-topic", as it is a useful answer to the posted question.
To elaborate on what Mr. Norway said, you can use MindTerm to give you a Unix terminal session from within a web browser. Then, using that session, you can use any mail client you choose that runs in a Unix terminal. These include Mutt, Pine, Elm, Mh, the Emacs mail modes, and the original Unix Mail.
HTML is not the only Web technology; Java applets have become fairly popular in recent years, and MindTerm is the single most useful one I've ever seen.
Mathias' new Palm Pilot oriented half-qwerty models appear to be left-handed only, but his original design worked just as well for people with only a right hand. The software version (which allows you to use a regular qwerty keyboard as if it was a half-qwerty) should do the trick for you quite nicely, although it does cost around $300. Check out halfkeyboard.com, as linked from the original post.
Although it doesn't have an integrated pointing device, you'd be no worse off than a two-handed person switching back and forth between a keyboard and a mouse.
As I said, the Windows and Mac versions are a little pricey for a pure software solution, but there's also an unofficial patch to the Linux kernel that does it for free
(try a web search; I don't have the link offhand).
The way you approach the problem will depend on whether you're primarily interested in telephony or audio. The concepts are similar, but the hardware and drivers come from different sources, and are optimized quite differently.
I can't provide any suggestions from the telephony side, but if you're interested in multichannel audio, check out the Advanced Linux Sound Architecture (ALSA) project. Unlike the older "OSS" sound drivers for Linux, ALSA has a significant amount of support for multichannel audio hardware. Although it is not a particularly mature project, it is reasonably well supported, and is used by nearly the entire professional audio community under Linux (such as it may be).
There is a version of AltaVista, available as
a commercial product, which would nicely meet
the needs you describe. It supports filtering
by category as well as keyword searches (along
with many other features).
Most chording keyboards (including the Twiddler) are seriously slow, and have a fairly steep learning curve. A researcher in Canada came up with a nifty one-handed keyboard system called Half Qwerty, which is nearly as fast as regular Qwerty and has almost zero learning curve. It leverages your existing skill at typing Qwerty combined with the fact that your hands are mirror images of one another.
This isn't really an answer to the question, but please bear with me... there are good reasons to run multiple soundcards simultaneously (for instance, recording each player in your band through a separate channel), but this doesn't sound like one of them. Instead, the standard approach is to mix the output of the three programs together in software, and send it to the speakers through a single card. This is exactly what ESD and the like were created to do.
I would think that in Linux, the API for creating non-disk-based-filesystems would be one of the most heavily used around, so that people could have filesystem-based access to databases, network protocols, the contents of archive files, etc. Unfortunately nobody seems to have done this at the operating system level; instead they waste their time building this functionality at the application level, like in Midnight Commander. I wonder why.
I'm personally going to try my hand at writing filesystem-based access to IMAP mail servers. My plan is to write a proxy that acts as an IMAP client and an NFS server, which can then be mounted using Linux or any other NFS-supporting operating system. By masquerading as an NFS server I can avoid using non-portable driver APIs. Now if only I had time to spend on coding it...
Indeed... the Slashdot article calls Analog "a small SciFi magazine", yet I was under the impression that it was just about the most preeminent periodical of the whole genre.
I'm not sure if there are any out-of-the-box MIDI Stage Control apps for Linux, but it would be fairly trivial to write one.
For a properly configured system, reading MIDI messages is simply a matter of opening/dev/midi (or/dev/midi## if you have multiple cards) calling read().
Seriously, it's that easy. If you understand the MIDI protocol (see www.midifarm.com or numerous other places), writing this kind of MIDI software for Linux is dead trivial. But my grandest creation, as history will tell,
I believe that the Sound Blaster PCI 128 uses the same chipset as the Sound Blaster Live, which is supported under ALSA (the newer, better, alternative set of sound and MIDI drivers for Linux). You can find several links to the ALSA site elsewhere in this forum. But my grandest creation, as history will tell,
I know "me too" replies are frowned upon, but this posting is too easy to miss.
Thus let me just say that Dave Phillip's "Linux Sound and Music" pages linked above are the worldwide clearing house for information on the subject of MIDI on Linux. If you can't find what you need there or linked from there, you'll most likely have to write it yourself! But my grandest creation, as history will tell,
I have had an original Music Quest "PC MIDI Card" MPU-401 compatible since they first came out, and have used it with Linux for a number of years now. What I've found is that for practical purposes, there is virtually zero difference between one of them and the MPU-401 "compatible" clones found in virtually every sound card these days.
The technical difference is that the Music Quest and the real Roland support an "intelligent mode", which performs buffering of outbound MIDI messages on the card. This was designed to allow the ancient PC-XT (which is what I first had mine plugged into) to handle MIDI even though it wasn't fast enough to do it very well purely in software.
Modern PC's, of course, are 100 to 150 times as fast, so they don't need intelligent mode. Instead, they can use the MPU-401 in raw mode, which does no buffering on the card; rather the card acts identically to a serial port, albeit with MIDI's odd transfer rate. This raw mode is what the Sound Blasters and the like support. It is also the only mode supported under Linux and Windows; only very esoteric old DOS software uses intelligent mode.
Hope this clears it up, and that you don't feel cheated for buying the Music Quest when it offers nothing more than a generic Sound Blaster for practical purposes.
Div. But my grandest creation, as history will tell,
Re:Ergo boards look like they melted in the sun.
on
Ergonomic Keyboards
·
· Score: 1
You can purchase brand new, authentic, click action keyboards with everything but the IBM sticker from Unicomp. Or so I've heard... I bought my four real IBM clackers used. But my grandest creation, as history will tell,
I don't agree that Logic, Cubase, Cakewalk, and other "professional quality" commercial sequencers actually suit the needs of their target users. I am in the target audience, have tried everything on the market, and find all of them to be very lacking.
The good news is that there are a whole crop of new sequencer projects coming from the open source community. In approximate order of vintage, these are Rosegarden, Jazz, Cantor, Gseq, Brahms (aka KooBase), Melys, and Muse. The bad news is that most of these are so busy trying to be clones of the leading commercial ones that they don't have any groundbreaking features of their own.
My own sequencer project "PEGS" is a long way from being ready for users. However, it is based on a very different premise from the leading commercial sequencers, and thus will be able to fill a unique, useful niche of its own when it's finally ready.
Slashdot Mirror
It's an interesting viewpoint, but I think you're placing too much emphasis on the physical aspects of a library. A library is just as much any collection of books or documents under communal ownership (your taxes make you a partial owner of your town's collection of books).
For a second, let's make the arguably unpleasant assumption that an artificial scarcity will be maintained amongst digital documents in the future (as foreshadowed by the current Ebook scandal). In such a situation, it would still be possible for groups of people to get together online and make their own private document sharing collective, in which the right for any one person to read a certain book at a certain time was passed around like a token. This would be functionally the same as a library today, but in the pure digital world.
This is obviously very different from the utopic vision of unlimited copies of digital documents being freely available to all. I don't mean to pass judgement one way or the other (since my opinion doesn't count for much anyway), but until such time as the world figures out a practical incentive for authors, composers, artists, etc to keep producing, then a library like this would not be such a bad idea.
But my grandest creation, as history will tell,
Some search engines have this capability built in, including the one which I work on professionally: AltaVista Enterprise Search.
But my grandest creation, as history will tell,
While Nielsen is fairly widely respected, this particular editorial of his is an absolute embarassment.
His criticisms fall into two categories:
None of these, of course, is a deficiency of the file format itself, and all are easily addressed. I'm not an apologist of any sort, but misrepresentations like this are not particularly useful.
But my grandest creation, as history will tell,
pushd and popd?
But my grandest creation, as history will tell,
If your script is the one capturing the form data, then it is also usually the one which generated the page with the form on it, so you can tell the browser to switch into whatever encoding you want (using the charset option on the Content-type HTTP header or placing it in an HTML META tag).
But my grandest creation, as history will tell,
Wrong. You're confusing the UCS-2, the encoding, with UCS, the character repertoire.
To spell it out, "Universal Character Set - Two Byte Encoding" (UCS-2) is one of many encodings which can represent the "Basic Multilingual Plane" (BMP) subset of the "Universal Character Set" (UCS) character repertoire.
But my grandest creation, as history will tell,
> Qt Embedded runs off a floppy if you wish (thought about trying to get that running on a 486 I had).
You underestimate the power of your old computer. My 486 runs XFree86 and KDE in its entirety. Sure it's slow, but it works just fine.
But my grandest creation, as history will tell,
I'm fond of the C64 version, myself. Actually, "addicted" would be a more accurate word. :-)
There was once an attempt to port it to DOS, but it was squashed by the copyright holder. You can probably still search for "Mule 386".
But my grandest creation, as history will tell,
John Gilmore wrote:
There was an old science fiction book by the name of A for Anything which explores the societal implications of just such a technology. It does not take anywhere's near as cheery a view as the EFF.
I haven't decided exactly where I stand on the issue, but Mr. Gilmore raises some good points.
Div.
But my grandest creation, as history will tell,
I would much sooner moderate this as "informative" than "off-topic", as it is a useful answer to the posted question.
To elaborate on what Mr. Norway said, you can use MindTerm to give you a Unix terminal session from within a web browser. Then, using that session, you can use any mail client you choose that runs in a Unix terminal. These include Mutt, Pine, Elm, Mh, the Emacs mail modes, and the original Unix Mail.
HTML is not the only Web technology; Java applets have become fairly popular in recent years, and MindTerm is the single most useful one I've ever seen.
But my grandest creation, as history will tell,
Mathias' new Palm Pilot oriented half-qwerty models appear to be left-handed only, but his original design worked just as well for people with only a right hand. The software version (which allows you to use a regular qwerty keyboard as if it was a half-qwerty) should do the trick for you quite nicely, although it does cost around $300. Check out halfkeyboard.com, as linked from the original post.
Although it doesn't have an integrated pointing device, you'd be no worse off than a two-handed person switching back and forth between a keyboard and a mouse.
As I said, the Windows and Mac versions are a little pricey for a pure software solution, but there's also an unofficial patch to the Linux kernel that does it for free (try a web search; I don't have the link offhand).
But my grandest creation, as history will tell,
The way you approach the problem will depend on whether you're primarily interested in telephony or audio. The concepts are similar, but the hardware and drivers come from different sources, and are optimized quite differently.
I can't provide any suggestions from the telephony side, but if you're interested in multichannel audio, check out the Advanced Linux Sound Architecture (ALSA) project. Unlike the older "OSS" sound drivers for Linux, ALSA has a significant amount of support for multichannel audio hardware. Although it is not a particularly mature project, it is reasonably well supported, and is used by nearly the entire professional audio community under Linux (such as it may be).
But my grandest creation, as history will tell,
Take a look at http://solutions.altavista.com under the heading AltaVista Search Engine 3.0.
A little disclaimer: I work as a developer for that product, so of course I'm biased; however, it really is very powerful.
But my grandest creation, as history will tell,
Most chording keyboards (including the Twiddler) are seriously slow, and have a fairly steep learning curve. A researcher in Canada came up with a nifty one-handed keyboard system called Half Qwerty, which is nearly as fast as regular Qwerty and has almost zero learning curve. It leverages your existing skill at typing Qwerty combined with the fact that your hands are mirror images of one another.
There's a patch to the Linux keyboard driver which implements it on top of a regular Qwerty keyboard.
Enjoy,
Div.
But my grandest creation, as history will tell,
According to Freshmeat, there is a program called "Dante" which might help. I haven't tried it myself, though.
But my grandest creation, as history will tell,
This isn't really an answer to the question, but please bear with me... there are good reasons to run multiple soundcards simultaneously (for instance, recording each player in your band through a separate channel), but this doesn't sound like one of them. Instead, the standard approach is to mix the output of the three programs together in software, and send it to the speakers through a single card. This is exactly what ESD and the like were created to do.
But my grandest creation, as history will tell,
I haven't used any of these myself, but they might work for you: a list of speech synthesis and analysis software for Linux.
But my grandest creation, as history will tell,
I would think that in Linux, the API for creating non-disk-based-filesystems would be one of the most heavily used around, so that people could have filesystem-based access to databases, network protocols, the contents of archive files, etc. Unfortunately nobody seems to have done this at the operating system level; instead they waste their time building this functionality at the application level, like in Midnight Commander. I wonder why.
I'm personally going to try my hand at writing filesystem-based access to IMAP mail servers. My plan is to write a proxy that acts as an IMAP client and an NFS server, which can then be mounted using Linux or any other NFS-supporting operating system. By masquerading as an NFS server I can avoid using non-portable driver APIs. Now if only I had time to spend on coding it...
Div.
But my grandest creation, as history will tell,
Indeed... the Slashdot article calls Analog "a small SciFi magazine", yet I was under the impression that it was just about the most preeminent periodical of the whole genre.
But my grandest creation, as history will tell,
I'm not sure if there are any out-of-the-box
/dev/midi /dev/midi## if you have multiple cards)
MIDI Stage Control apps for Linux, but it would
be fairly trivial to write one.
For a properly configured system, reading MIDI
messages is simply a matter of opening
(or
calling read().
Seriously, it's that easy. If you understand
the MIDI protocol (see www.midifarm.com or
numerous other places), writing this kind of
MIDI software for Linux is dead trivial.
But my grandest creation, as history will tell,
I believe that the Sound Blaster PCI 128 uses the same chipset as the Sound Blaster Live, which is supported under ALSA (the newer, better, alternative set of sound and MIDI drivers for Linux). You can find several links to the ALSA site elsewhere in this forum.
But my grandest creation, as history will tell,
I know "me too" replies are frowned upon, but
this posting is too easy to miss.
Thus let me just say that Dave Phillip's
"Linux Sound and Music" pages linked above
are the worldwide clearing house for
information on the subject of MIDI on Linux.
If you can't find what you need there or
linked from there, you'll most likely have
to write it yourself!
But my grandest creation, as history will tell,
I have had an original Music Quest "PC MIDI Card" MPU-401 compatible
since they first came out, and have used it with Linux for a number
of years now. What I've found is that for practical purposes, there
is virtually zero difference between one of them and the MPU-401
"compatible" clones found in virtually every sound card these days.
The technical difference is that the Music Quest and the real Roland
support an "intelligent mode", which performs buffering of outbound
MIDI messages on the card. This was designed to allow the ancient
PC-XT (which is what I first had mine plugged into) to handle MIDI
even though it wasn't fast enough to do it very well purely in
software.
Modern PC's, of course, are 100 to 150 times as fast, so they don't
need intelligent mode. Instead, they can use the MPU-401 in raw
mode, which does no buffering on the card; rather the card acts
identically to a serial port, albeit with MIDI's odd transfer rate.
This raw mode is what the Sound Blasters and the like support. It
is also the only mode supported under Linux and Windows; only very
esoteric old DOS software uses intelligent mode.
Hope this clears it up, and that you don't feel cheated for buying
the Music Quest when it offers nothing more than a generic
Sound Blaster for practical purposes.
Div.
But my grandest creation, as history will tell,
You can purchase brand new, authentic, click action keyboards with everything but the IBM sticker from Unicomp. Or so I've heard... I bought my four real IBM clackers used.
But my grandest creation, as history will tell,
I don't agree that Logic, Cubase, Cakewalk, and other "professional quality" commercial sequencers actually suit the needs of their target users. I am in the target audience, have tried everything on the market, and find all of them to be very lacking.
The good news is that there are a whole crop of new sequencer projects coming from the open source community. In approximate order of vintage, these are Rosegarden, Jazz, Cantor, Gseq, Brahms (aka KooBase), Melys, and Muse. The bad news is that most of these are so busy trying to be clones of the leading commercial ones that they don't have any groundbreaking features of their own.
My own sequencer project "PEGS" is a long way from being ready for users. However, it is based on a very different premise from the leading commercial sequencers, and thus will be able to fill a unique, useful niche of its own when it's finally ready.
-- Div.
But my grandest creation, as history will tell,