We actually once tried bringing live audience into our Shoutcast radio. We encoded the Shoutcast stream by piping esdmon | lame... | shout..., and thus could mix multiple music sources (from EsounD-supporting players) as well as speech.
For the remote speaker, we tried running esdrec and then esdmon | lame... | nc broadcaster-host:12345.
For the broadcaster, we joined the remote speaker into the stream by: nc -l 12345 | mpg123 -s - | esdcat.
The only problem was - it had a latency of about 5 seconds:)
From my experience, the forms are submitted with the page's encoding. If the page containing the form is a UTF-8 encoded page, the content would be submitted as UTF-8 encoded. Simply set your HTTP header "Content-Type" to the right encoding (e.g. Content-Type: text/html; charset=UTF-8).
Internet Explorer also tends to convert posted messages with characters which don't fit into the current encoding (e.g. Russian text, while the page is in ISO-8859-1 encoding) into numerical HTML entities (Ӓ...) which are the character's position in the UCS-2 (Unicode) table.
Are there other ways to know what the browser meant? I'm not sure.
You're not a geek. Nothing we can do about it. You won't understand.
Just notice those are news for geeks. We are the ones who don't explore computers for pure application purposes. We enjoy our little addictions. There are plenty of business-man directed reviews on CNET -- here we strive to find others who appreciate the same things we appreciate as geeks. And part of our impression of a product is it's geek "coolness" factor.
What about things like win2k logons/remote access?
Samba has offered working as a PDC and offering Login-into-domain functionality (tested by me on Win95 / Win98 boxes) for ages. This is what defines what Unix user you create files as on the Samba share etc. and accordingly your permissions.
Lately, I've been realizing more and more how the concept of the Internet is going to hell.
For example, here in Israel, the most-used link we have is an optical connection to the US. Nobody cares of connection anywhere else, and even ISPs which have connections to Europe (e.g. Barak ITC which represents Global One in Israel) doesn't offer the European link to the common users. About connections to our neighboring countries, there's very little to talk about, since they're both mostly technically undeveloped and aren't in very friendly diplomatic relationships with us, to say it mildly. So it ends up that we route via the US to reach Turkey or the far east.
In case of a war, which is sadly something more likely in our region, there would be just one point of failure.
Of course, one of the leading ISPs, NetVision, seems to have relatively broadband satellite links which might be the solution.
Recently, we've been reading lots of articles and opinions on what path and "business plan" we should take, or Linux will never make it and "we'll lose". There were articles linked from Slashdot, and a friend of mine also commented recently that we "need to simplify Linux or else nobody will use it".
So, I'd rather just answer him cynically "yes, Linux is losing, and soon gonna be erradicated, and we're so damn scared of losing the market", while the reality speaks for itself.
The desktop developers don't ignore the UI. KDE and GNOME do attempt running UI research and usability tests just like the big guys, but they're always doing it for the hack value, not cause we owe something to someone and want to win hordes of users.
The idea is to hack and have fun, not to market our "Linux" thing. We don't owe it to anymore.
GTK+ features a very nice feature of keyboard-bindings-on-the-fly. Simply move your mouse over the menu item (without clicking it), press the desired key combination - and there you have it. Wish to erase the combination? Move the mouse above the menu item, and press Del.
Actually yes, this can be bad publicity, since, unlike the Linux kernel, many Linux applications are as unstable as their Windows counterparts, simply because they're in development.
<p>I even recently had to answer a newbie who said <i>"Why does Netscape and Mozilla keep crashing all the time? Wasn't Linux supposed to be more stable, advanced etc? IE never crashed on me as much."</i>. Obviously, he didn't draw the line between kernel and third-party applications.
At the Internet World (Israel) 2000 expo, I've seen an Israeli company selling their so-called "MagicCard", which was supposed to allow recovery from any problem, be it modified settings, format C: or a total secure wipe of the disk (with Norton Utilities).
After I inqueried more, they claimed it works on all operating systems, and via a very special compression method manages to backup your up to 26GB hard drive into 300MB of data.
Later on, I asked a friend who runs a school lab with those things. According to him, the thing simply hangs the computer once it has no space to store the session no more.
Oh yeah, BitBoys -- first they fail to release Pyramid3D, then Glaze3D... and now? "Xtreme Bandwidth Architecture"? What the hell are those guys doing?
To rephrase some E2 node whose title I can no longer remember, E2 would be what a future generations would like to open to get an idea of who we were.
Currently E2 is a fair mix of facts, rants, GTKY (Getting-To-Know-You) polls and inside jokes -- but thinking of it, it represents us much better than just plain facts. And those future generations wouldn't wanna read a dry encyclopedia either. Thus, a pile of dry facts can be as useless as a pile of GTKY polls.
"No, bollocks, I want as many people as possible to see my page/buy my books/read my posts, and if my designers can't be arsed to make a page that the biggest possible audience can see, then that's my problem. It's nothing to do with my customers."
Unfortunately, most webmasters nowadays would say "We run best on Internet Explorer. Upgrade.". We meet this frequently while advocating for the Mozilla / Konqueror teams, finding sites which do not work and reporting their owners (sometimes very large companies).
I had to go through this myself lately too, when I was running a long series of mails and replies with the Ozon.ru staff (a big Russian book/video store) about trivial HTML bugs on their page. Instead of realizing I have a clue, they simply told me my "Netscape 6" isn't yet supported, so I should use Internet Explorer... or email them my order details manually!
So much for "addressing the biggest possible audience"...
<p>Although nowadays it's popular to hate Enlightenment, bash it's speed and fanciness, Enlightenment for me was more like a proof of concept, an attempt to achieve what's not directly achievable by X11. It was also a major eye candy in Linux advocasy demos (well, till MacOS X arrived, with the "swallowing taskbar":).
<p>I'm talking about the rippling desktop, zooming workspaces, transparent terminals (eTerm). All of that stuff was done by hacks around X, without any dedicated effects API from X -- and they managed to do it.
<p>Now, OpenGL can offer those effects with real hardware acceleration - scaling, alpha blending, just the stuff all those effects need.
Even worse, correct me if I am wrong, it seems that Netscape and IExplorer never did sort out whether to use the "Layer" tag or the (retracted) ILayer tag, or ids and styles to create dynamic html.
Netscape Navigator 4.0 did support and tags, though they were discouraged and it did support CSS-P (positionable stylesheets), almost as well as Internet Explorer 4.0. The API wasn't a standard (document.layer[menu].top = '5px') and you couldn't change CSS properties (not position, but things like background color) on the fly.
Internet Explorer 4.0 and up only supported W3C-based stylesheets solution for positioning content ("ids and styles"), but the API to accesss it via the JScript wasn't a standard (document.all.menu.style.top = '5px').
Internet Explorer 5.0 retained the document.all compatibility and also added a JScript API compliant with the W3C ECMAScript bindings for DOM specs (for example, each object has a getElementById method, so you'd be writing: document.getElementById('menu').style.top = '5px')
Mozilla supports only the W3C way of both creating positionable objects and using them in JavaScript.
Thus, when you design a DHTML site:
if internet_explorer_4 then internet_explorer_specific_code;
else if netscape_navigator_4 then netscape_specific_code;
Netscape Navigator 4 has so many quirks though, you might end up debugging it for days, so judge for yourself if you really want it.
Tip: A nice way to shorten variations of the code would be to detect browsers, and then fill up an array with references to the "layers" you intend to work with. For Moz/IE, add the 'layer.style' object for each "layer". For Nescape, add the 'layer' object for each "layer". Then you could simply use: array[menu].top = '5px'.
Err...which part of HTML4 doesn't Netscape 6 support?
<p>BiDi (Bidirectional languages, e.g. Arabic, Hebrew) support is missing, atleast until the IBMBIDI patches (made by IBM) finally get merged into the mainstream build.
Nor do I ever touch swap despite running simultaneous instances of gimp, emacs, multiple xterms compiling, editing and debugging as well as the usual assortment of servers and then some.
How much RAM do you have, so that you have "enough of it" and never hit swap?:)
Developing DHTML for Mozilla CAN only be done in a W3C compliant form. While now this might annoy you, it's a really good decision for the long run not to provide back compatibility for NS4/IE proprietary DOMs.
Anyhow:
- Mozilla, Internet Explorer 5 = document.getElementById('quack').style
- Internet Explorer 4 =
document.all.quack.style
- Netscape Navigator 4 =
document.layers.quack
The getElementById and getElementByName functions are defined by the W3C document ECMA bindings for DOM2. Look it up.
Dreamweaver authors should have researched better before making such claims.
ICANN, instead of sorting and generalizing content by generic TLDs, is trying to come up with TLDs to please the site owners and domain registrars. Those of course prefer the trendy cool names, like.ebiz.
Why the E? Because 'electronic' is the buzzword today? It's on the web, so it's already electronic. And why "biZ"? Why using a buzzword term, instead of a short for the real word "business"?
Why.sex? Or.xxx? What about sites which contain violence or any other sort of "entertainment" suitable only for adults? Why not make it a generic ".adult" or ".adu"?
As to ".bz", the threat must be ignored. Country TLDs were given for a reason - to represent a country. Bending the rules by having enough money and tricking the system is a bad thing to be accepted. I think ICANN should make a reverse claim, that people might confuse people into thinking ".bz" sites represent Belizze (since it's the official meaning of the TLD), while in fact those are from a bought out legacy TLD.
If the goal is to have '.cool', '.sucks', '.notsucks', '.gnu' etc., why not drop the whole thing and have just the domain names? (ala Compuserve's GO MICROSOFT)
Indeed. Thawte's been the most friendly 'major company' I've seen yet.
The Web of Trust: instead of Verisign personal certificates, only costing money and making no assumptions about identity, unlike Thawte, which issues such certificates for free, following a Web of Trust scheme - never just to make money
Very friendly to the open source people, supports PGP and open source web servers, even back then when Verisign refused to know any non-commercial SSL implementation for Apache
Now, would Verisign sell Thawte certificates when Verisign's root CA expires at 1/1/2000 on Navigator 4.5 and lower? I wonder if this was their intention...
Slashdot Mirror
For the remote speaker, we tried running esdrec and then esdmon | lame ... | nc broadcaster-host:12345.
For the broadcaster, we joined the remote speaker into the stream by: nc -l 12345 | mpg123 -s - | esdcat.
The only problem was - it had a latency of about 5 seconds :)
Internet Explorer also tends to convert posted messages with characters which don't fit into the current encoding (e.g. Russian text, while the page is in ISO-8859-1 encoding) into numerical HTML entities (Ӓ...) which are the character's position in the UCS-2 (Unicode) table.
Are there other ways to know what the browser meant? I'm not sure.
http://www.iglu.org.il/linux_report.pdf
If you want more, just get Abiword and do it yourself!
S/PDIF for AudioPCI
You're not a geek. Nothing we can do about it. You won't understand.
Just notice those are news for geeks. We are the ones who don't explore computers for pure application purposes. We enjoy our little addictions. There are plenty of business-man directed reviews on CNET -- here we strive to find others who appreciate the same things we appreciate as geeks. And part of our impression of a product is it's geek "coolness" factor.
Samba has offered working as a PDC and offering Login-into-domain functionality (tested by me on Win95 / Win98 boxes) for ages. This is what defines what Unix user you create files as on the Samba share etc. and accordingly your permissions.
For example, here in Israel, the most-used link we have is an optical connection to the US. Nobody cares of connection anywhere else, and even ISPs which have connections to Europe (e.g. Barak ITC which represents Global One in Israel) doesn't offer the European link to the common users. About connections to our neighboring countries, there's very little to talk about, since they're both mostly technically undeveloped and aren't in very friendly diplomatic relationships with us, to say it mildly. So it ends up that we route via the US to reach Turkey or the far east.
In case of a war, which is sadly something more likely in our region, there would be just one point of failure.
Of course, one of the leading ISPs, NetVision, seems to have relatively broadband satellite links which might be the solution.
And the copycat howmanywouldittake.com too.
Recently, we've been reading lots of articles and opinions on what path and "business plan" we should take, or Linux will never make it and "we'll lose". There were articles linked from Slashdot, and a friend of mine also commented recently that we "need to simplify Linux or else nobody will use it".
So, I'd rather just answer him cynically "yes, Linux is losing, and soon gonna be erradicated, and we're so damn scared of losing the market", while the reality speaks for itself.
The desktop developers don't ignore the UI. KDE and GNOME do attempt running UI research and usability tests just like the big guys, but they're always doing it for the hack value, not cause we owe something to someone and want to win hordes of users.
The idea is to hack and have fun, not to market our "Linux" thing. We don't owe it to anymore.
GTK+ features a very nice feature of keyboard-bindings-on-the-fly. Simply move your mouse over the menu item (without clicking it), press the desired key combination - and there you have it. Wish to erase the combination? Move the mouse above the menu item, and press Del.
Actually yes, this can be bad publicity, since, unlike the Linux kernel, many Linux applications are as unstable as their Windows counterparts, simply because they're in development.
<p>I even recently had to answer a newbie who said <i>"Why does Netscape and Mozilla keep crashing all the time? Wasn't Linux supposed to be more stable, advanced etc? IE never crashed on me as much."</i>. Obviously, he didn't draw the line between kernel and third-party applications.
At the Internet World (Israel) 2000 expo, I've seen an Israeli company selling their so-called "MagicCard", which was supposed to allow recovery from any problem, be it modified settings, format C: or a total secure wipe of the disk (with Norton Utilities).
After I inqueried more, they claimed it works on all operating systems, and via a very special compression method manages to backup your up to 26GB hard drive into 300MB of data.
Later on, I asked a friend who runs a school lab with those things. According to him, the thing simply hangs the computer once it has no space to store the session no more.
Oh yeah, BitBoys -- first they fail to release Pyramid3D, then Glaze3D ... and now? "Xtreme Bandwidth Architecture"? What the hell are those guys doing?
Currently E2 is a fair mix of facts, rants, GTKY (Getting-To-Know-You) polls and inside jokes -- but thinking of it, it represents us much better than just plain facts. And those future generations wouldn't wanna read a dry encyclopedia either. Thus, a pile of dry facts can be as useless as a pile of GTKY polls.
Unfortunately, most webmasters nowadays would say "We run best on Internet Explorer. Upgrade.". We meet this frequently while advocating for the Mozilla / Konqueror teams, finding sites which do not work and reporting their owners (sometimes very large companies).
I had to go through this myself lately too, when I was running a long series of mails and replies with the Ozon.ru staff (a big Russian book/video store) about trivial HTML bugs on their page. Instead of realizing I have a clue, they simply told me my "Netscape 6" isn't yet supported, so I should use Internet Explorer ... or email them my order details manually!
So much for "addressing the biggest possible audience" ...
What Unicode has to do with fonts? You still need Hebrew glyphs. Unicode is about data representation. Fortunatelly, decent fonts are not hard to get.
No, it's Qt 2, which does represent it's strings internally in Unicode. But without Right-to-Left rendering, nor input. (wait for Qt 3)
That's obvious!
:).
<p>Although nowadays it's popular to hate Enlightenment, bash it's speed and fanciness, Enlightenment for me was more like a proof of concept, an attempt to achieve what's not directly achievable by X11. It was also a major eye candy in Linux advocasy demos (well, till MacOS X arrived, with the "swallowing taskbar"
<p>I'm talking about the rippling desktop, zooming workspaces, transparent terminals (eTerm). All of that stuff was done by hacks around X, without any dedicated effects API from X -- and they managed to do it.
<p>Now, OpenGL can offer those effects with real hardware acceleration - scaling, alpha blending, just the stuff all those effects need.
Netscape Navigator 4.0 did support and tags, though they were discouraged and it did support CSS-P (positionable stylesheets), almost as well as Internet Explorer 4.0. The API wasn't a standard (document.layer[menu].top = '5px') and you couldn't change CSS properties (not position, but things like background color) on the fly.
Internet Explorer 4.0 and up only supported W3C-based stylesheets solution for positioning content ("ids and styles"), but the API to accesss it via the JScript wasn't a standard (document.all.menu.style.top = '5px').
Internet Explorer 5.0 retained the document.all compatibility and also added a JScript API compliant with the W3C ECMAScript bindings for DOM specs (for example, each object has a getElementById method, so you'd be writing: document.getElementById('menu').style.top = '5px')
Mozilla supports only the W3C way of both creating positionable objects and using them in JavaScript.
Thus, when you design a DHTML site:
Netscape Navigator 4 has so many quirks though, you might end up debugging it for days, so judge for yourself if you really want it.
Tip: A nice way to shorten variations of the code would be to detect browsers, and then fill up an array with references to the "layers" you intend to work with. For Moz/IE, add the 'layer.style' object for each "layer". For Nescape, add the 'layer' object for each "layer". Then you could simply use: array[menu].top = '5px'.
Err...which part of HTML4 doesn't Netscape 6 support?
<p>BiDi (Bidirectional languages, e.g. Arabic, Hebrew) support is missing, atleast until the IBMBIDI patches (made by IBM) finally get merged into the mainstream build.
How much RAM do you have, so that you have "enough of it" and never hit swap? :)
Try this patch for the NVIDIA-KERNEL package. http://www.linuxgames.com/misc/patch-nvdriver-2.4. 0-test11-2
Developing DHTML for Mozilla CAN only be done in a W3C compliant form. While now this might annoy you, it's a really good decision for the long run not to provide back compatibility for NS4/IE proprietary DOMs.
Anyhow:
- Mozilla, Internet Explorer 5 = document.getElementById('quack').style
- Internet Explorer 4 =
document.all.quack.style
- Netscape Navigator 4 =
document.layers.quack
The getElementById and getElementByName functions are defined by the W3C document ECMA bindings for DOM2. Look it up.
Dreamweaver authors should have researched better before making such claims.
How's about compiling SDL with aalib and then:
SDL_VIDEODRIVER=aalib aviplay pr0n.avi
ICANN, instead of sorting and generalizing content by generic TLDs, is trying to come up with TLDs to please the site owners and domain registrars. Those of course prefer the trendy cool names, like .ebiz.
.sex? Or .xxx? What about sites which contain violence or any other sort of "entertainment" suitable only for adults? Why not make it a generic ".adult" or ".adu"?
Why the E? Because 'electronic' is the buzzword today? It's on the web, so it's already electronic. And why "biZ"? Why using a buzzword term, instead of a short for the real word "business"?
Why
As to ".bz", the threat must be ignored. Country TLDs were given for a reason - to represent a country. Bending the rules by having enough money and tricking the system is a bad thing to be accepted. I think ICANN should make a reverse claim, that people might confuse people into thinking ".bz" sites represent Belizze (since it's the official meaning of the TLD), while in fact those are from a bought out legacy TLD.
If the goal is to have '.cool', '.sucks', '.notsucks', '.gnu' etc., why not drop the whole thing and have just the domain names? (ala Compuserve's GO MICROSOFT)
- The Web of Trust: instead of Verisign personal certificates, only costing money and making no assumptions about identity, unlike Thawte, which issues such certificates for free, following a Web of Trust scheme - never just to make money
- Very friendly to the open source people, supports PGP and open source web servers, even back then when Verisign refused to know any non-commercial SSL implementation for Apache
- They offer free IRC-based technical support
- They link and suggest Fortify on their site
- They have the best privacy policy I've ever seen - read it - you'll like it!
Now, would Verisign sell Thawte certificates when Verisign's root CA expires at 1/1/2000 on Navigator 4.5 and lower? I wonder if this was their intention...