Microsoft will need to be broken into more than 3 parts, and they will all need new names. I'd like to suggest:
Baby Bill #1: Nanosoft Baby Bill #2: Bill's Software Repair Baby Bill #3: Necrosoft Baby Bill #4: Steve's Drivethru Benchmarks Inc. Baby Bill #5: Zombiesoft Baby Bill #6: Black Hat Software Baby Bill #7: Microwas --
Sure, they could all work together--but simple game theory predicts they won't.
Simple game theory isn't good enough for me, especially when the playing field isn't level.
With such a demonstrably bad actor as Microsoft both kinds of remedy are needed. To make a short proof by example: what about Kerberos? Even with the OS division hived off, Kerberos is still at risk from attack by NT servers and W98ME clients running "extended" versions of the protocol (read: intentionally compatible only with Microsoft products, and, to make matters worse, secret). Even with the IE division separate from the division that sells IIS (yes, they'll have to SELL it!), a conduct remedy is necessary to prevent those two Micro-Microsofts from cooperating, to their mutual benefit, by continuing to implement and support non-standard versions of web protocols, known only to the two of them. Etc. etc, there are many such strategems available for defeating the intent of remedies if they are purely structural. All these strategems need to be anticipated and addressed.
By the same token conduct remedies alone are insufficient. This was more than adequately demonstrated by the way Microsoft (read: Bill) was able to weasel around the earlier consent decree on product tying.
Both kinds of remedy are obviously needed. If any weaker approach is applied we'll just see the whole sordid affair playing out again 3 years down the road, wasting everybody's time and money, including Microsoft's.
And don't think any of this is unfair to Microsoft - you live by the sword you die by the sword. None of this is going to harm Microsoft to nearly the same extent they harmed their competitors, their allies (e.g., IBM), and legions of poor, befuddled consumers (who have come to believe that crashing is just part of the normal behaviour you expect from a computer and that rebooting is part of installing an application).
At worst the Micro-Microsofts will simply have to compete like normal companies from now on. That means, hmm, writing business plans like anyone else, getting capital from the same places everybody else gets it instead of extorting it from witless consumers, making deals that are beneficial to both parties and don't turn into the kinds of rape scenes we've seen in the past, going hat in hand to the bank, an so on. In other words, normal healthy capitalism. --
Please give me docs that I can sit down with outside and leaf through at my leisure, with no need to be anywhere near a computer.
And please give me online docs in html format (at minimum!) - no PDF or postscript please, unless this is also accomanied by html.
By far the most important version is the online, hypertext, searchable version. Whenever there's a tradeoff to be made for cost reasons, favor the online docs.
On the other hand, when I pay big money for a software product I *expect* printed docs and if I don't get them I probably won't be back a second time. --
I still use pkzip/unzip constantly, mostly for backup under Dos/Windows (my day job:P). It's nice to work with a compression program so reliable you can use it for backup.
PKzip is, by the way, enormously faster than winzip in my experience. --
Tragedy is an important component of being genious.
It may be common for tragedy and genius to be linked, but it sure isn't an absolute requirement. I give as a counterexample J.S. Bach, who according to all the information we have, was a very happy man with a good family life (and what a family!) More unhappy geniuses to look to this man as an example.
Also, consider that misery is hardly limited to the intellectually endowed. It's just that, with genius, it hurts you more and you are more able to communicate your hurt to others. --
This may be seen as a feature, not a bug by some;-) However there's no denying the fact that the web needs animated pictures, if only so advertisers can annoy us so much that we buy their product:-o
The fact is, the lack of animation in PNG is going to be the biggest obstacle to eliminating GIF. Browser support is no longer really a problem, as soon as Mozilla ships - as Microsoft provides enough PNG support to get by (though both have trouble with alpha right now, a sexy feature that gif never supported anyway). What to do? What we really should have done is be ready with MNG, the animated network graphics format, but it just didn't happen - I suppose it's still not ready, for some reason completely incomprehsible to me, and even if it were it's not supported in Mozilla (yet:) and certainly not in in IE.
What then to do? It's possible to use javascript to animate PNGs, but (1) not everybody surfs with Javascript enabled (2) that's not exactly a trivial change to make to a web page. Ideally we'd just want to replaces GIFs by equivalent PNGs but that's just not possible with animated GIFs.
At the risk of sounding like a broken record, what to do??? I sure don't know the answer to this, but I know something needs to be done. Not only does Unisys suck with it's abusive royalty policy and sleazy submarine tactic, GIFs suck too: (1) only 256 colors (2) poor compression (3) very difficult to scale dynamically.
I thought "hey, lets see if I can understand this NOW!" Guess what? It's still spaghetti code. I still can't unstand a stick of it, other then the PRINTFs and SCANFs. That's it. And I got a 98% in the class.
Hang in there. In programming as in any other highly developed speciality it takes a while to get from where you have to start as a beginner to where the pros are working. And also, don't give up. Just because you don't understand the first function in a listing (that's actually pretty normal) doesn't mean you can't go somewhere else in the listing and start with something you *do* understand. Hint: start at the bottom and read up. Start with "main":-)
Another factor is that you often have to have a pretty good grounding in the application area the code is written for before you can have a clue what it's doing, so if you're reading cryto code, go check out some crypto sites and maybe get a cryto bood. Same goes for security. Read the howto's and get an idea what the apps are trying to do. Good documentation helps a *lot* too, and while it may not always be included in the source code;-) you can often find it in other places, for example, the web site the code came from, or news group archives, or the linuxdoc project, etc.
And then there's just outright bad coding style... it's often not your fault that you can't read something right away. Poorly chosen variable names, unecessarily redundant code;-) badly structured code and use of obscure library functions all contribute to code being difficult to read.
But the main thing is just experience. It takes a minimum of 3 years to get to a professional level of knowlege in c++ and until you get there, you'll have to pick and choose the code you work on and you'll have to work hard to understand typical pieces of production code. Also consider that you'll get there faster if you start with a language other than c++ - say, Java or Python, because the concepts you need to learn are much more clearly expressed in those languages with fewer obscure and distracting features.
In the meantime, don't worry, there are still plenty of eyeballs out here that can read and understand this stuff:-) --
If Linus Torvalds subscribed to that philosophy, we wouldn't have linux.
This is not going to work... if it appeared three, maybe four years ago, they did stand a chance of success...
You don't know that. All it takes is one big game to adopt it, say Quake or Unreal, for the next cross-platform release, and an API like this will take off. The absolute minimum that can happen is a lot of good work can be done that can be incorporated into Loki's SDL, which has already taken off IMHO.
Game publishers just stare blankly when you mention the possibility for non-Windows versions of a game.
Yes, and IT managers reacted the same way to Linux 2 years ago. My how things change. The Linux gamers base is now probably sitting around 10 million and doubling every year, conservatively. That's enough to get the attention of people who matter. Have you been in a retail shop lately? Linux games are on the shelves:) Actually, a lot more of them than I'd have expected at this point. I tend to think the Linux shelf space will actually increase faster than the linux game community itself for a couple of reasons: (1) linux users tend to be a little richer than windows users (2) retailers recognize the value of front-loading in a rapidly expanding market.
...missing from the list... NVIDIA [they'll show up soon enough] Apple [them too I'd think] Creative [ah, they're not going to miss the party if it starts to rock] SEGA, Sony and Nintendo [does that really matter?]
Look who's on the list: 3dfx, 3Dlabs, ATI, Compaq, Discreet, Evans & Sutherland, IBM, Intel, S3, and SGI
Something good is going to come out of this, there's no question about it. --
Ahhh, yes the only way to enjoy art is to pay for it. Guess the idea of public art, street preformers, public art houses, and art accessability projects are bunk. Thanks for pointing this out Lars, otherwise I would have never have known that art is to be bought, and not enjoyed
Excellent point. Till now, performing art has enjoyed the special distinction of being an unending source of revenue for the artist, or at least, the artist's contract-holders. While non-performing artists have generally been rewarded with lump-sum payments for their work from collectors, museums, etc, deriving only a small part of their income from licensing of prints, photographic reproductions and so on.
I don't see that the fine arts world is any the worse for this, and we certainly don't seem to be short of artists. Should we now admit that what is happening is a fundamental, unstoppable, change in the way the business of music is done?
As a musician myself, I don't see that these changes are bad. What is really being torn down is the system of manufacturing stars - entirely a creation of the music industry. If there is an endless supply of bands just as talented as metallica, and believe me, there is, then what you'll see is the recording industry no longer being able to control the vast bulk of the supply of music by limiting the number of newly minted stars. What's going to happen? Obviously, once the chokehold on supply is broken, prices will fall. That's exactly what's happening, although in all the confusion, it's sometimes hard to see that.
What we're really seeing here is the beginning of a real flowering of art for art's sake, in much the same way as the current exponential growth of code hacking for code hacking's sake. It's not necessary to drive a Benz to be a functioning member of society as a musician. There are still many ways to be paid for the work, and to be paid well for great work, but one of them - the one that supports the recording industry's monopoly on talent and supply - is coming to an end. --
Hopefully, this is a good time and place to broadcast a message to skin developers:-)
The 6 basic navigation buttons I want in every skin are:
1) back 2) forward 3) up (shortens the url by one segment) 4) reload 5) stop 6) home
In that order. Mozilla has it *almost* right, but not quite (the up button is missing).
I also want to be able to define multiple "home" buttons and have them appear beside the usual home button, in every sense equal, and not have my own buttons relegated to a space-consuming "personal tool bar".
While I'm going on about my wishes... what about all that space to the left of the "help" item? I want to be able to put the location box there, saving a whole bunch of valuable vertical real estate.
Hmm - just one more wish today - when I minimize a tool bar, how nice it would be if the space to the right of the little tab were transparent, winning me back a tiny bit more vertical real-estate, instead of opaque, empty space.
Thanks in advance, skin hackers, for taking this seriously:-) --
What's up with this registering and asking me about the size of my mother's army boots just so I can get SO5.2 BETA!!!
StarOffice is, unfortunately, in the clutches of an evil corporation. Oh sure, there are far eviler corporations out there (I could name one right now:) but Sun is certainly no angel, especially with respect to LInux which they perceive - correctly - as a threat to Solaris. Course they would be a lot smarter to leverage Linux instead of doing the subtle sandbagging-with-a-smile they engage in now, but what the heck, that's just my opinion.
In the short term, Sun owning StarOffice is good for Linux - it will speed Linux takeup on the desktop. But in the long run, we basically have to kiss StarOffice goodbye as an open-source program. That's probably bad.
Or maybe it's not. Actually, Abiword is really coming along and once it gets fleshed out a little more, I think I'll prefer using it by a ***wide*** margin of some of the other bloatware office products. KOffice is really promising too, and Gnome supplies a few pieces of the puzzle (gnumeric is already darn good and it's getting better every week). --
Try it, you'll like it. It's my impression that Mozilla is considerably ahead of Netscape 6 prerel at this point, and that's as it should be. Although I have to admit, this Netscape build is impressively stable considering... (2nd slashdot post, 4-5 windows open for an hour on a limited-memory machine, and STILL ALIVE!!!:) --
Absolutely agreed. Checking out the Mozilla home page, you see that while the number of "outside" developers has increased only slowly, the number of checkins from outside has skyrocketed. This exactly mirrors my own intentions - I've always promised myself that I'd actually start hacking the code and contributing better features (in the sense that *I* will like them more:) only after the commercial version of Netscape is forked off.
The reasons for this are simple.
1) Don't even think about trying to get a new feature, as opposed to a bug fix into the source tree while the Netscape team is on the straight stretch towards a commercial release.
2) Now that Netscape is separate, Mozilla is a LOT more "ours".
I think I'm hardly the only developer that feels this way. I'd look for the number of outside developers to double in the next month.
BTW, for anyone that's interested, if you want to hack on Mozilla you need a gig of free disk space, and good fast processor - I'd suggest at least 400 MHz, and any current version of Linux. Downloading the source takes about 30 mins on a 56K modem and the initial build takes about 40 mins. In contrast to other reports I've seen, I've never had a *single* nightly tree fail to build on Linux. It's actually pretty darn easy to get into, compared to some source installs I've done.
(Posted with Netscape 6 pre-release of course:) --
How exactly is Freenet going to protect itself from:
Spammers... filling the servers with ads for porn sites and get-rich-quick schemes
Flooders... since it's free, grabbing all the bandwidth just to be annoying
Pranksters... posting bogus data just to devalue the real data
Slanderers, pornographers, warez kiddies and other bottomfeeders of the net doing their sleazy thing and giving the service a bad name.
Don't get me wrong, I'm all for Freenet! In fact I'm already on the mailing list and I'm joining the project. But... what about those tough problems? --
one thing that is negligected when talks of the result of this case are mentioned are finacal ones; not just how it will affect MS's finances, but the stock market. Everyone knows that most high tech stocks, especially the IPOs of late, are riding a bubble, and at some point, that bubble *has* to burst. I think both Judge Jackson and most market people realize that even with a ruling on MS that is in accordance with the law and anti-monopoly tactics, it WILL hurt the market.
The only market it will hurt is the market for Microsoft shares. Think of how many Linux and independent software vendor shares are going to get a shot in the arm from a ruling that promises to bell the cat, er, um, tame the shrew, hmmm, cut the gordian knot, ahhhh, you know what I mean. It sure wouldn't hurt to be holding some Redhat and Corel when the conclusions of law come down.
...When that stock bubble breaks, it probably will not take down the whole market, but it will ripple through the current booming economy badly.
If, as you suggest, there is some tough medicine to be taken, isn't it best to take it when the economy is strongest?
In any event, the market as a whole will turn up if anything, the stronger the ruling, the more up.
No, all Jackson's efforts to achieve settlement have nothing at all to do with the stock market. They have everything to do with (1) trying to achieve effective rememdies without a long draw-out supreme court appeal (2) bringing down a rock-solid judgement that doesn't stand a chance of being overturned on appeal. Showing that every effort was made in good faith to achieve a settlement, and that the defendant did not reciprocate in good faith is just part of building that airtight judgement. No judge likes to be reversed on appeal and Jackson is doing everything he can to avoid that.
Full journalling of everything, but without the journal?
Instant crash recovery without having to replay a log?
Backward compatible with ext2, supports all ext2 features
Similar speed to ext2 - hardly any penalty for being failsafe
A *small* patch to ext2 and only about 20k extra runtime code
Well, that's what I've been designing the last couple of years and coding the last few months - it's called Tux2 and I'm going to announce it soon on linux-kernel. It still has a couple of bugs, apparently races, but now I'm going to play the open source card and get help scratching those itches.
If you're interested in helping out, email me at: phillips (at) bonn-fries (dot) net
This is just the Microsoft way of making you think you are getting something when you are really not. Opening windows source isn't going to benefit too many people. The source is a big mess and too large to sort through without spending a great deal of time on it. Think about it. In linux... recompile the kernal source the way you want it. In windows... change it and beg microsoft to put it in?!?
If Microsoft releases source code, it has to **build** and it has to create the **same** binaries as the released ones. That's a no-brainer, but, we still might have to do some explaining to convince the DOJ that that's the only way it will work. Furthermore, it has to be source code for **all** versions of Windows, current and future, or it just plain won't work. And finally, it has to be the source for **all* of Windows, not just the kernel.
To put it simply: "the source code, the whole windows source code, and nothing but buildable source code"
It's got a tiny text window and a minimal set of features, but it's terminally cute and rugged as heck. I've splattered it across the pavement about 5 times by having it fall out of my pocket when I'm riding my bike... and no damage. I had to clean the battery contacts once. I will definitely buy one again - one that's even smaller (although the 768 is *still* one of the smallest after being out almost 2 years). I just which I could program it myself though, because I know I could do a better job. *sigh*. Will hardware makers ever get it? All they have to do is open the specs, and we'll make their phone twice as good as it already is.
Drop the price of a CD to $1.50 and put 55% of that straight into the pocket of the artists. This model works. At that price I'd buy a CD 9 time out of 10 instead of downloading it, finding space for it and/or buring a CD at probably even higher cost.
Everybody wins here except the RIA (sob sob, weep weep)
1. If I install Red Hat on a certain set of Machines (Gateway 2000 P5-120 & P5-166's) X fails to start. Every time, after a clean install, it detects the card right, but doesn't work, just bombs back to the command prompt.
2. X works fine in Mandrake, BUT, the ethernet cards do not! They work fine in Red Hat, however. Again, they're detected, and as far as I can tell are using the same drivers, but they don't work.
3. Corel Linux works fine with both of these things, right out of the box. Yes, I know Corel is based on Debian, not Red Hat. I'm just saying that it works.:)
The X problems are X problems, not Linux problems per se. That said, I've been there too and, while initially frustrating, it doesn't take too much digging to get it sorted out. The two problems I've run into frequently are: (1) resolution/color depth not configured properly (Redhat's Xconfigurator will fix this, and if you really want to you can fix it by hand in the XF86Config file) (2) No entry for your hostname in resolv.conf. Now, I think it's just plain weird that X doesn't start because of (2) but the fix is pretty obvious (just enter you hostname/ip into resolv.conf).
Linux would in general would be a ***whole lot*** easier to install/use if XFree weren't such a strange beast. Oh well, that will change in time - see yesterday's thread that mentioned MicroWindows and NanoGui. Help is on its way.
As far as ethernet cards go - they're pretty hard to break. Probably you just don't have the proper kernel module installed. That's a matter of doing "insmod yourethermodule". Normally this is done automatically, but some install programs just don't get it right. It may also be possible that the module for your particular ethernet wasn't compiled in the kernel that was shipped (maybe because the card is too new?) and what you have to do is install/compile the kernel source. Which isn't that hard and it's pretty satisfying once you've figured it out.
The thing is, if you have similar problems in Windows - and believe me, you will - you basically don't have any options but: reboot; reinstall the driver; reboot; reinstall Windows; reboot; reboot; reboot; FDISK; repeat.:-(
As far as I know, Matrox has the most open specs so far (with the possible exception of S3) and the G400 is a kick-ass 3D card that compares favorably to NVidia both in terms of price and performance. This should be a no-brainer.
Too bad, NVidia, you're losing it *just like 3D fx did*. Do the words "those who can't remember the past are doomed to repeat it" mean anything to you?????
Go and download a nightly build and give it a try. If it crashes or has annoying bugs, report them and download a new one a few days later.
It does crash, but you now have to work at it. I'm doing about 1/2 my browsing with Mozilla now, which I'd say is a pretty significant benchmark. Interesting thing: I've run across about 5-6 sites that you *need* Mozilla for (any page ending with.xml for example). I guess there's going to be a lot more of that as time goes by:-)
Mozilla is quite a bit smaller than Netscape 7 and it now loads faster, though not yet nearly as fast as Konqueror or Opera.)
I can post to Slashdot with reasonable confidence I won't lose my text:-) and that's a big improvement - though I still wouldn't try anything too fancy when I've got a form up (e.g., hitting back and forward and expecting to still have your field contents intact.)
There are minor formatting problems that vary a lot from build to build. I downloaded my first skin yesterday (aphrodite - looks good). Memory usage has improved a lot, but it's still got a ways to go. The bugs are down to a short enough list that it's worth keeping track of them and sending them in.
I can remember when the browser's scrolling had a one second delay and the whole UI was sticky. Now it's really fast at least on Win32.
It's fast on Linux too.
I'm absolutely positive that Mozilla will be a success.
There's no question about that. I also have a feeling that the serious hacking on Mozilla is only going to start after Netscape 6 forks off officially. Then I think you'll see a lot more latitude for *real* changes and improvements to the user interface - beyond just skins and standards - because that narrow focus on just getting the thing out the door will be gone, and the netscape team won't be so protective about the feature set (I hope I hope I hope).
There's also other key features of COM which make building HUGE operating systems like Windows 2000 and Office possible.
This isn't a flame, but frankly I'd rather go slowly, get it right, and NOT wind up with a huge operating system like Windows 200. ***But Still*** have all the functionality. In fact, have more functionality, and more reliable by virtue of being built on a more solid, tighter base.
For me, this means one thing: build it on XML. Toss both COM *and* CORBA: both of these are first-attempt kinds of thingies. One suffers from various kinds of tumours and the other suffers from terminal bloat. Time to back up, do it over and do it right. Gnome ppl: listen up - CORBA just isn't appropriate for client-based componentization. KDE ppl: listen up too - kparts is an ok concept and DCOP is a more lightweight transport than IIOP (*grunt* *groan*) but XML-based interface definition and XML-based interapp communication is going to make both of them obsolete. What that means is... be prepared to rewrite the component/interapp layer yet again for KDE 3.
Sometimes it takes a few tries, but when you get it right, it's right. COM isn't right. Neither is CORBA. When something in an OS is not right, you're morally bound to toss it (really). XML is (((probably right))) so let's get some prototypes happening, um, yesterday.:-)
Slashdot Mirror
Microsoft will need to be broken into more than 3 parts, and they will all need new names. I'd like to suggest:
Baby Bill #1: Nanosoft
Baby Bill #2: Bill's Software Repair
Baby Bill #3: Necrosoft
Baby Bill #4: Steve's Drivethru Benchmarks Inc.
Baby Bill #5: Zombiesoft
Baby Bill #6: Black Hat Software
Baby Bill #7: Microwas
--
Sure, they could all work together--but simple game theory predicts they won't.
Simple game theory isn't good enough for me, especially when the playing field isn't level.
With such a demonstrably bad actor as Microsoft both kinds of remedy are needed. To make a short proof by example: what about Kerberos? Even with the OS division hived off, Kerberos is still at risk from attack by NT servers and W98ME clients running "extended" versions of the protocol (read: intentionally compatible only with Microsoft products, and, to make matters worse, secret). Even with the IE division separate from the division that sells IIS (yes, they'll have to SELL it!), a conduct remedy is necessary to prevent those two Micro-Microsofts from cooperating, to their mutual benefit, by continuing to implement and support non-standard versions of web protocols, known only to the two of them. Etc. etc, there are many such strategems available for defeating the intent of remedies if they are purely structural. All these strategems need to be anticipated and addressed.
By the same token conduct remedies alone are insufficient. This was more than adequately demonstrated by the way Microsoft (read: Bill) was able to weasel around the earlier consent decree on product tying.
Both kinds of remedy are obviously needed. If any weaker approach is applied we'll just see the whole sordid affair playing out again 3 years down the road, wasting everybody's time and money, including Microsoft's.
And don't think any of this is unfair to Microsoft - you live by the sword you die by the sword. None of this is going to harm Microsoft to nearly the same extent they harmed their competitors, their allies (e.g., IBM), and legions of poor, befuddled consumers (who have come to believe that crashing is just part of the normal behaviour you expect from a computer and that rebooting is part of installing an application).
At worst the Micro-Microsofts will simply have to compete like normal companies from now on. That means, hmm, writing business plans like anyone else, getting capital from the same places everybody else gets it instead of extorting it from witless consumers, making deals that are beneficial to both parties and don't turn into the kinds of rape scenes we've seen in the past, going hat in hand to the bank, an so on. In other words, normal healthy capitalism.
--
Please give me docs that I can sit down with outside and leaf through at my leisure, with no need to be anywhere near a computer.
And please give me online docs in html format (at minimum!) - no PDF or postscript please, unless this is also accomanied by html.
By far the most important version is the online, hypertext, searchable version. Whenever there's a tradeoff to be made for cost reasons, favor the online docs.
On the other hand, when I pay big money for a software product I *expect* printed docs and if I don't get them I probably won't be back a second time.
--
I still use pkzip/unzip constantly, mostly for backup under Dos/Windows (my day job :P). It's nice to work with a compression program so reliable you can use it for backup.
PKzip is, by the way, enormously faster than winzip in my experience.
--
Tragedy is an important component of being genious.
It may be common for tragedy and genius to be linked, but it sure isn't an absolute requirement. I give as a counterexample J.S. Bach, who according to all the information we have, was a very happy man with a good family life (and what a family!) More unhappy geniuses to look to this man as an example.
Also, consider that misery is hardly limited to the intellectually endowed. It's just that, with genius, it hurts you more and you are more able to communicate your hurt to others.
--
This may be seen as a feature, not a bug by some ;-) However there's no denying the fact that the web needs animated pictures, if only so advertisers can annoy us so much that we buy their product :-o
The fact is, the lack of animation in PNG is going to be the biggest obstacle to eliminating GIF. Browser support is no longer really a problem, as soon as Mozilla ships - as Microsoft provides enough PNG support to get by (though both have trouble with alpha right now, a sexy feature that gif never supported anyway). What to do? What we really should have done is be ready with MNG, the animated network graphics format, but it just didn't happen - I suppose it's still not ready, for some reason completely incomprehsible to me, and even if it were it's not supported in Mozilla (yet:) and certainly not in in IE.
What then to do? It's possible to use javascript to animate PNGs, but (1) not everybody surfs with Javascript enabled (2) that's not exactly a trivial change to make to a web page. Ideally we'd just want to replaces GIFs by equivalent PNGs but that's just not possible with animated GIFs.
At the risk of sounding like a broken record, what to do??? I sure don't know the answer to this, but I know something needs to be done. Not only does Unisys suck with it's abusive royalty policy and sleazy submarine tactic, GIFs suck too: (1) only 256 colors (2) poor compression (3) very difficult to scale dynamically.
Suggestions?
--
I thought "hey, lets see if I can understand this NOW!" Guess what? It's still spaghetti code. I still can't unstand a stick of it, other then the PRINTFs and SCANFs. That's it. And I got a 98% in the class.
:-)
;-) you can often find it in other places, for example, the web site the code came from, or news group archives, or the linuxdoc project, etc.
;-) badly structured code and use of obscure library functions all contribute to code being difficult to read.
:-)
Hang in there. In programming as in any other highly developed speciality it takes a while to get from where you have to start as a beginner to where the pros are working. And also, don't give up. Just because you don't understand the first function in a listing (that's actually pretty normal) doesn't mean you can't go somewhere else in the listing and start with something you *do* understand. Hint: start at the bottom and read up. Start with "main"
Another factor is that you often have to have a pretty good grounding in the application area the code is written for before you can have a clue what it's doing, so if you're reading cryto code, go check out some crypto sites and maybe get a cryto bood. Same goes for security. Read the howto's and get an idea what the apps are trying to do. Good documentation helps a *lot* too, and while it may not always be included in the source code
And then there's just outright bad coding style... it's often not your fault that you can't read something right away. Poorly chosen variable names, unecessarily redundant code
But the main thing is just experience. It takes a minimum of 3 years to get to a professional level of knowlege in c++ and until you get there, you'll have to pick and choose the code you work on and you'll have to work hard to understand typical pieces of production code. Also consider that you'll get there faster if you start with a language other than c++ - say, Java or Python, because the concepts you need to learn are much more clearly expressed in those languages with fewer obscure and distracting features.
In the meantime, don't worry, there are still plenty of eyeballs out here that can read and understand this stuff
--
If Linus Torvalds subscribed to that philosophy, we wouldn't have linux.
:) Actually, a lot more of them than I'd have expected at this point. I tend to think the Linux shelf space will actually increase faster than the linux game community itself for a couple of reasons: (1) linux users tend to be a little richer than windows users (2) retailers recognize the value of front-loading in a rapidly expanding market.
...missing from the list... NVIDIA [they'll show up soon enough] Apple [them too I'd think] Creative [ah, they're not going to miss the party if it starts to rock] SEGA, Sony and Nintendo [does that really matter?]
This is not going to work... if it appeared three, maybe four years ago, they did stand a chance of success...
You don't know that. All it takes is one big game to adopt it, say Quake or Unreal, for the next cross-platform release, and an API like this will take off. The absolute minimum that can happen is a lot of good work can be done that can be incorporated into Loki's SDL, which has already taken off IMHO.
Game publishers just stare blankly when you mention the possibility for non-Windows versions of a game.
Yes, and IT managers reacted the same way to Linux 2 years ago. My how things change. The Linux gamers base is now probably sitting around 10 million and doubling every year, conservatively. That's enough to get the attention of people who matter. Have you been in a retail shop lately? Linux games are on the shelves
Look who's on the list: 3dfx, 3Dlabs, ATI, Compaq, Discreet, Evans & Sutherland, IBM, Intel, S3, and SGI
Something good is going to come out of this, there's no question about it.
--
Ahhh, yes the only way to enjoy art is to pay for it. Guess the idea of public art, street preformers, public art houses, and art accessability projects are bunk. Thanks for pointing this out Lars, otherwise I would have never have known that art is to be bought, and not enjoyed
Excellent point. Till now, performing art has enjoyed the special distinction of being an unending source of revenue for the artist, or at least, the artist's contract-holders. While non-performing artists have generally been rewarded with lump-sum payments for their work from collectors, museums, etc, deriving only a small part of their income from licensing of prints, photographic reproductions and so on.
I don't see that the fine arts world is any the worse for this, and we certainly don't seem to be short of artists. Should we now admit that what is happening is a fundamental, unstoppable, change in the way the business of music is done?
As a musician myself, I don't see that these changes are bad. What is really being torn down is the system of manufacturing stars - entirely a creation of the music industry. If there is an endless supply of bands just as talented as metallica, and believe me, there is, then what you'll see is the recording industry no longer being able to control the vast bulk of the supply of music by limiting the number of newly minted stars. What's going to happen? Obviously, once the chokehold on supply is broken, prices will fall. That's exactly what's happening, although in all the confusion, it's sometimes hard to see that.
What we're really seeing here is the beginning of a real flowering of art for art's sake, in much the same way as the current exponential growth of code hacking for code hacking's sake. It's not necessary to drive a Benz to be a functioning member of society as a musician. There are still many ways to be paid for the work, and to be paid well for great work, but one of them - the one that supports the recording industry's monopoly on talent and supply - is coming to an end.
--
Hopefully, this is a good time and place to broadcast a message to skin developers :-)
:-)
The 6 basic navigation buttons I want in every skin are:
1) back
2) forward
3) up (shortens the url by one segment)
4) reload
5) stop
6) home
In that order. Mozilla has it *almost* right, but not quite (the up button is missing).
I also want to be able to define multiple "home" buttons and have them appear beside the usual home button, in every sense equal, and not have my own buttons relegated to a space-consuming "personal tool bar".
While I'm going on about my wishes... what about all that space to the left of the "help" item? I want to be able to put the location box there, saving a whole bunch of valuable vertical real estate.
Hmm - just one more wish today - when I minimize a tool bar, how nice it would be if the space to the right of the little tab were transparent, winning me back a tiny bit more vertical real-estate, instead of opaque, empty space.
Thanks in advance, skin hackers, for taking this seriously
--
What's up with this registering and asking me about the size of my mother's army boots just so I can get SO5.2 BETA!!!
StarOffice is, unfortunately, in the clutches of an evil corporation. Oh sure, there are far eviler corporations out there (I could name one right now:) but Sun is certainly no angel, especially with respect to LInux which they perceive - correctly - as a threat to Solaris. Course they would be a lot smarter to leverage Linux instead of doing the subtle sandbagging-with-a-smile they engage in now, but what the heck, that's just my opinion.
In the short term, Sun owning StarOffice is good for Linux - it will speed Linux takeup on the desktop. But in the long run, we basically have to kiss StarOffice goodbye as an open-source program. That's probably bad.
Or maybe it's not. Actually, Abiword is really coming along and once it gets fleshed out a little more, I think I'll prefer using it by a ***wide*** margin of some of the other bloatware office products. KOffice is really promising too, and Gnome supplies a few pieces of the puzzle (gnumeric is already darn good and it's getting better every week).
--
Try it, you'll like it. It's my impression that Mozilla is considerably ahead of Netscape 6 prerel at this point, and that's as it should be. Although I have to admit, this Netscape build is impressively stable considering... (2nd slashdot post, 4-5 windows open for an hour on a limited-memory machine, and STILL ALIVE!!!:)
--
Absolutely agreed. Checking out the Mozilla home page, you see that while the number of "outside" developers has increased only slowly, the number of checkins from outside has skyrocketed. This exactly mirrors my own intentions - I've always promised myself that I'd actually start hacking the code and contributing better features (in the sense that *I* will like them more:) only after the commercial version of Netscape is forked off.
The reasons for this are simple.
1) Don't even think about trying to get a new feature, as opposed to a bug fix into the source tree while the Netscape team is on the straight stretch towards a commercial release.
2) Now that Netscape is separate, Mozilla is a LOT more "ours".
I think I'm hardly the only developer that feels this way. I'd look for the number of outside developers to double in the next month.
BTW, for anyone that's interested, if you want to hack on Mozilla you need a gig of free disk space, and good fast processor - I'd suggest at least 400 MHz, and any current version of Linux. Downloading the source takes about 30 mins on a 56K modem and the initial build takes about 40 mins. In contrast to other reports I've seen, I've never had a *single* nightly tree fail to build on Linux. It's actually pretty darn easy to get into, compared to some source installs I've done.
(Posted with Netscape 6 pre-release of course:)
--
ftp://ftp.netscape.com/pub/netscape6/english/6_PR1 /unix/linux22/ne tscap e-v600pr1.x86-unknown-linux2.2.tar.gz
--
Don't get me wrong, I'm all for Freenet! In fact I'm already on the mailing list and I'm joining the project. But... what about those tough problems?
--
one thing that is negligected when talks of the result of this case are mentioned are finacal ones; not just how it will affect MS's finances, but the stock market. Everyone knows that most high tech stocks, especially the IPOs of late, are riding a bubble, and at some point, that bubble *has* to burst. I think both Judge Jackson and most market people realize that even with a ruling on MS that is in accordance with the law and anti-monopoly tactics, it WILL hurt the market.
...When that stock bubble breaks, it probably will not take down the whole market, but it will ripple through the current booming economy badly.
The only market it will hurt is the market for Microsoft shares. Think of how many Linux and independent software vendor shares are going to get a shot in the arm from a ruling that promises to bell the cat, er, um, tame the shrew, hmmm, cut the gordian knot, ahhhh, you know what I mean. It sure wouldn't hurt to be holding some Redhat and Corel when the conclusions of law come down.
If, as you suggest, there is some tough medicine to be taken, isn't it best to take it when the economy is strongest?
In any event, the market as a whole will turn up if anything, the stronger the ruling, the more up.
No, all Jackson's efforts to achieve settlement have nothing at all to do with the stock market. They have everything to do with (1) trying to achieve effective rememdies without a long draw-out supreme court appeal (2) bringing down a rock-solid judgement that doesn't stand a chance of being overturned on appeal. Showing that every effort was made in good faith to achieve a settlement, and that the defendant did not reciprocate in good faith is just part of building that airtight judgement. No judge likes to be reversed on appeal and Jackson is doing everything he can to avoid that.
Full journalling of everything, but without the journal?
Instant crash recovery without having to replay a log?
Backward compatible with ext2, supports all ext2 features
Similar speed to ext2 - hardly any penalty for being failsafe
A *small* patch to ext2 and only about 20k extra runtime code
:-)
Well, that's what I've been designing the last couple of years and coding the last few months - it's called Tux2 and I'm going to announce it soon on linux-kernel. It still has a couple of bugs, apparently races, but now I'm going to play the open source card and get help scratching those itches.
If you're interested in helping out, email me at: phillips (at) bonn-fries (dot) net
Programmers only, please
This is just the Microsoft way of making you think you are getting something when you are really not. Opening windows source isn't going to benefit too many people. The source is a big mess and too large to sort through without spending a great deal of time on it. Think about it. In linux... recompile the kernal source the way you want it. In windows... change it and beg microsoft to put it in?!?
If Microsoft releases source code, it has to **build** and it has to create the **same** binaries as the released ones. That's a no-brainer, but, we still might have to do some explaining to convince the DOJ that that's the only way it will work. Furthermore, it has to be source code for **all** versions of Windows, current and future, or it just plain won't work. And finally, it has to be the source for **all* of Windows, not just the kernel.
To put it simply: "the source code, the whole windows source code, and nothing but buildable source code"
It's got a tiny text window and a minimal set of features, but it's terminally cute and rugged as heck. I've splattered it across the pavement about 5 times by having it fall out of my pocket when I'm riding my bike... and no damage. I had to clean the battery contacts once. I will definitely buy one again - one that's even smaller (although the 768 is *still* one of the smallest after being out almost 2 years). I just which I could program it myself though, because I know I could do a better job. *sigh*. Will hardware makers ever get it? All they have to do is open the specs, and we'll make their phone twice as good as it already is.
Drop the price of a CD to $1.50 and put 55% of that straight into the pocket of the artists. This model works. At that price I'd buy a CD 9 time out of 10 instead of downloading it, finding space for it and/or buring a CD at probably even higher cost.
Everybody wins here except the RIA (sob sob, weep weep)
Offtopic? Which moderator screwed up here??? That was fuunnnnnnnieeeeeeeee.
1. If I install Red Hat on a certain set of Machines (Gateway 2000 P5-120 & P5-166's) X fails to start. Every time, after a clean install, it detects the card right, but doesn't work, just bombs back to the command prompt.
:)
:-(
2. X works fine in Mandrake, BUT, the ethernet cards do not! They work fine in Red Hat, however. Again, they're detected, and as far as I can tell are using the same drivers, but they don't work.
3. Corel Linux works fine with both of these things, right out of the box. Yes, I know Corel is based on Debian, not Red Hat. I'm just saying that it works.
The X problems are X problems, not Linux problems per se. That said, I've been there too and, while initially frustrating, it doesn't take too much digging to get it sorted out. The two problems I've run into frequently are: (1) resolution/color depth not configured properly (Redhat's Xconfigurator will fix this, and if you really want to you can fix it by hand in the XF86Config file) (2) No entry for your hostname in resolv.conf. Now, I think it's just plain weird that X doesn't start because of (2) but the fix is pretty obvious (just enter you hostname/ip into resolv.conf).
Linux would in general would be a ***whole lot*** easier to install/use if XFree weren't such a strange beast. Oh well, that will change in time - see yesterday's thread that mentioned MicroWindows and NanoGui. Help is on its way.
As far as ethernet cards go - they're pretty hard to break. Probably you just don't have the proper kernel module installed. That's a matter of doing "insmod yourethermodule". Normally this is done automatically, but some install programs just don't get it right. It may also be possible that the module for your particular ethernet wasn't compiled in the kernel that was shipped (maybe because the card is too new?) and what you have to do is install/compile the kernel source. Which isn't that hard and it's pretty satisfying once you've figured it out.
The thing is, if you have similar problems in Windows - and believe me, you will - you basically don't have any options but: reboot; reinstall the driver; reboot; reinstall Windows; reboot; reboot; reboot; FDISK; repeat.
As far as I know, Matrox has the most open specs so far (with the possible exception of S3) and the G400 is a kick-ass 3D card that compares favorably to NVidia both in terms of price and performance. This should be a no-brainer.
Too bad, NVidia, you're losing it *just like 3D fx did*. Do the words "those who can't remember the past are doomed to repeat it" mean anything to you?????
Go and download a nightly build and give it a try. If it crashes or has annoying bugs, report them and download a new one a few days later.
.xml for example). I guess there's going to be a lot more of that as time goes by :-)
:-) and that's a big improvement - though I still wouldn't try anything too fancy when I've got a form up (e.g., hitting back and forward and expecting to still have your field contents intact.)
It does crash, but you now have to work at it. I'm doing about 1/2 my browsing with Mozilla now, which I'd say is a pretty significant benchmark. Interesting thing: I've run across about 5-6 sites that you *need* Mozilla for (any page ending with
Mozilla is quite a bit smaller than Netscape 7 and it now loads faster, though not yet nearly as fast as Konqueror or Opera.)
I can post to Slashdot with reasonable confidence I won't lose my text
There are minor formatting problems that vary a lot from build to build. I downloaded my first skin yesterday (aphrodite - looks good). Memory usage has improved a lot, but it's still got a ways to go. The bugs are down to a short enough list that it's worth keeping track of them and sending them in.
I can remember when the browser's scrolling had a one second delay and the whole UI was sticky. Now it's really fast at least on Win32.
It's fast on Linux too.
I'm absolutely positive that Mozilla will be a success.
There's no question about that. I also have a feeling that the serious hacking on Mozilla is only going to start after Netscape 6 forks off officially. Then I think you'll see a lot more latitude for *real* changes and improvements to the user interface - beyond just skins and standards - because that narrow focus on just getting the thing out the door will be gone, and the netscape team won't be so protective about the feature set (I hope I hope I hope).
There's also other key features of COM which make building HUGE operating systems like Windows 2000 and Office possible.
:-)
This isn't a flame, but frankly I'd rather go slowly, get it right, and NOT wind up with a huge operating system like Windows 200. ***But Still*** have all the functionality. In fact, have more functionality, and more reliable by virtue of being built on a more solid, tighter base.
For me, this means one thing: build it on XML. Toss both COM *and* CORBA: both of these are first-attempt kinds of thingies. One suffers from various kinds of tumours and the other suffers from terminal bloat. Time to back up, do it over and do it right. Gnome ppl: listen up - CORBA just isn't appropriate for client-based componentization. KDE ppl: listen up too - kparts is an ok concept and DCOP is a more lightweight transport than IIOP (*grunt* *groan*) but XML-based interface definition and XML-based interapp communication is going to make both of them obsolete. What that means is... be prepared to rewrite the component/interapp layer yet again for KDE 3.
Sometimes it takes a few tries, but when you get it right, it's right. COM isn't right. Neither is CORBA. When something in an OS is not right, you're morally bound to toss it (really). XML is (((probably right))) so let's get some prototypes happening, um, yesterday.