I recall a couple of years back, surrounding the release of Windows 95, an interview with Bill Gates on ABC's 20/20. The end of the interview closed with the interviewer being practically pushed out of Bill's office with Bill screaming at his secretary/assistant "I said five minutes. He had his five minutes, now he has to go. Mister? I've given you five minutes." It sounded a lot like something a young child throwing a temper tantrum after he'd already been cleaning his room "Forever". It makes you wonder if the Microsoft corporate hierarchy doesn't consist of people yelling downwards at the serfs who churn out thousands of lines of code and tens of pages of documentation a day.
How many times have you heard a Macintosh zealot give Windows users a hard time because nearly every Windows user-interface feature appeared in MacOS first? It is just me, or does the new Finder look a lot like the hideously-slow Microsoft "Active Desktop"? Everything's contained in one window (that looks strangely like a web browser). Big, hideous buttons/icons for everything. Lots of extraneous eye candy. As an ex-Mac user, I always loved the way Apple kept their UI clean. Things were kept simple (IE: fast and well thought-out) yet elegant. Even the idea of "exploding" windows was a great UI clue to the user to answer "What made that pop up?". But this? Have they any idea how many long-time Mac users they will alienate by tearing up the user interface that was even present on the old Lisa and Apple (not Mac) IIgs? The site mentions extending the "one window" metaphor to the entire operating system because "you can only interact with one application at a time". Doesn't this sound a lot like the days before the Multifinder? Wasn't one of System 7's greatest achievements the ability to multitask visually? The idea of BSD running behind it all sounds excellent, as Apple's OS has had a lot of holes in its design (seeming inability to use the hardware MMUs correctly, for one thing). However, they've taken what seems to be an obvious leap backwards in UI design by eliminating the clarity of pupose that every visible item in the Mac UI used to have. As a user, and as someone who has to provide support for users, I certainly hope there's an "Old Finder Look-and-Feel" option in "General Controls". I mean, come on. Apple's incarnation of the "Desktop" metaphor has propogated to OS/2, Windows, and who-knows-how-many window managers. Isn't tossing all this out the window just "thinking" a bit "too different"?
Dr.Scheme, Rice University'simplementation of MzScheme (also known as Rice Scheme) is now up to version 101. Back when I took COMP 212 there (the class that uses Dr.Scheme exclusively), the software was at version 53. That was only Fall 1998, so perhaps they've confused "build number" with "version number".:)
Now, these are just my thoughts on the recent annoying postings and what I'd do with them... take them with a grain of salt or ignore them.... I just happen to think this would be a good solution, and don't quite know where to post it. Automated zapping/banning is simply a bad idea. There's no easy way to tell an abusive post programatically. Take, for example, Microsoft's "Trash our Win2K Server" contest a few months back. Microsoft had a guestbook that was initially wide open (like Slashdot before moderation). Then, it has a curseword filter. As soon as the immatures realized this, they found creative ways around it (insertion of periods and spaces between letters). The filter also bleeped legitimate words, the most ironic of which being "competition". This isn't a problem to be dealt with by machines. Banning IP addresses for X hours/days/weeks/etc. is just as error-prone. Countless people have mentioned the problems with dial-up pools, IP masquerading, and falsified IPs. There are simply too may ways to prevent the wrong (IE: non-inflammatory) people from speaking. What I propose (and andover.net probably won't like this) is that someone should identify these posters. How? It's not too difficult, really. First, some background on myself. I work as the senior administrator for an Internet Service Provider in central Texas. I speak from the experience of that job, but I do not speak on their behalf. The organization I work for maintains a log of which user connects to which IP at what time. It maintains a complementary log of when the user's session is ended. This information can be used to pin down a specific dial-up account to an IP address. Most modern ISP s use a similar system (It's known as RADIUS accounting), as it's become a de facto standard (developed by Lucent. There's an RFC for it). This solves only half the problem, and from here is where it gets ugly. ISPs won't give out names or contact information unless something is harmful or illegal. In Texas, we're actually not allowed to give out that information without a subpoena, AFAIK. Thus, you have to have some grounds to recoup losses. I propose an "Acceptable Use Policy". It should state the policies of slashdot.org, including what sorts of posts are considered unacceptable. It should also state that posting several repetitious (esp. like today when some were over several kilobytes!) such messages may cause the IP to be traced and the site administrator to be contacted. A link to this policy should, of course appear whenever someone would be permitted to post. I am not an attorney, so I don't know what these sorts of things should say to be iron-clad, but I do think that the potent threat of action would be enough to deter such posts to an extent. Rob, et al, I know the last thing you guys want to be are policemen. However, take a look at SegFault. The abuse got so bad there that posting and write-ins were disabled completely. As I (and likely others) see it, one of the following solutions will play out:
The annoying bastards continue until the signal-to-noise ratio here is unbearable (we only have a finite number of moderator points)
People turn up their thresholds to the point that if you're not Linus Torvalds, Bruce Perens, or RMS, your post won't even get read.
We continue to bitch about the idiots
Of course, another obvious answer would be to implement a "cancel post" feature accessible to a few super-moderators that would effectively "kill" a post by setting its score to -2 or -infinity. That's an awful lot of power, though, and I think it would sit horribly with the Slashdot community at large. I know this is long and somewhat off-topic (although this article beautifully demonstrates the need for action), but I think it's important. Please post your ideas as a follow-up. It's time to take our forum back! PS: If someone sees this post as worthwhile and knows of a better place to post/send it, please let me know. Do the obvious thing to my posted email address to contact me. Spam and flamage not welcome, but coherent comments always are.
Kinda generic uh ? We can help if you give a little more details.
I don't know exactly what happened. Everything went fine, until the progress indicator showed that 40% or so of the files had been copied. It stopped for a bit, and then the indicator jumped to 100%, and promptly closed down without creating any icons on the Start Menu. When I checked the bin/ directory of the install-tree, it was barren, except for the batch-file that kick-starts JBuilder Foundation. It's probably because I was running Win2k on my test-machine; however, I really don't see why it wouldn't run on Windows 98 or 95 w/ Sun's JRE.
Have you tried ? The installer and JBuilder themselves have no native code. The JDK is a different story but then if you're writing about Java you probably already have a JDK for the platforms you mentioned.
No, I was just reading the requirements from your site. If it'll run on the stock JDK, that's excellent. Why don't you put this on your site under a "it'll probably work, but we don't officially support it" heading? I'd not mind giving it a shot on one of my S/Linux machines or an SGI. I just didn't want to take the time to download it, if it probably wasn't going to run on what I had. I'd bet that you could get a much larger following if word got out that it worked on most Unixes with a Java 2-compliant JRE.
The installer is written in Java with a loader written in Unix shell. It should run on any Unix system. KDE is not required.
Excellent! That's what I love to see! I'll give it a whirl sometime soon, then, as I've been looking for a decent Java IDE for Unix. Hopefully it's as cool as BCW 4.5 was... "back in the day".
I have to admit that the start of my hate of RedHat comes from my experiences with it on the SPARC platform. RedHat 5.something (not their first release on SPARC, but their first in quite a while) shipped so broken out-of-box that I ended up having to cross-compile half of the libraries on a Solaris machine just so that I could reliably compile. Eventually, I got the machine running (keep in mind... I'm not a newbie. I've been an administrator on Solaris, SunOS 4, IRIX 5, IRIX 6, and Linux for several years), and I've tried to poke the configuration as little as possible. Hopefully, over this Christmas holiday, I'll have time to put either Debian or Solaris 7 on the machine. As far as other distributions go, I'm a big fan of Slackware on the x86 platforms. Installation's a snap, it installs only what you ask for. I've walked people through installations of RedHat where they specifically turned off any X support, and RH decided that some subcomponent of some package needed X for a frontend. So, RH installs X, and X needs this and that and something else, and fonts, and blah-blah-blah. I've briefly played with Debian on the SPARC. Its installation routine is definitely easier than RedHat's, although using dselect later to add packages can be confusing if you don't read the README files first. There's also the lack of precompiled software for Debian, as RedHat's RPM is very popular. Although, being a SPARC user, I can rarely find S/Linux RPMs, anyway. I guess my main gripe with RedHat is that they're showing their "true self" as a traditional American corporation, instead of an organization dedicated to furthering the sensible development of an operating environment. They tend to suffer from shovelware syndrome, have a tendency to ship OSes with beta versions of the C runtime library, and have very-much a black box mentality when customizing the OS. This all well-and-good for a new user (or is it?) who doesn't know his way around the OS, but it's damned frustrating to a Unix veteran. I'd been trying to port Slackware to the SPARC platform. However, Patrick Volkerding wasn't interested in blessing my work, as SuSE had burned him before in such a deal. That, combined with my workload caused me to all but abandon the probject. I dunno... people bitch about Slackware because it doesn't behave like a System V Unix. I bitch about Redhat because it behaves like "RedHat OS" instead of like a Unix-like operating environment. I guess it all boils down to the fact that RH 5.0 was utter crap, and they lost me as a customer because of it's instability on both SPARC and Intel platforms. They have this get-out-the-door-and-ship-erratas-later mentatlity that has all but killed HP-UX's credibility and continues to plague Microsoft. I tend to like things that work well out-of-box. If I have to recompile sendmail or ftpd because of some new exploit, that's perfectly acceptable. However, if I can't compile out of box because libc.a or the compiler is broken, that's simply not acceptable, IMHO.
Actually, I'm complaining about the fact that a supposedly "pure-Java" application is platform-specific. Isn't this what landed Microsoft in the fire a few months back? If you gave a car, I wouldn't bitch at the colour. I'd bitch if it only took left-exits onto prime-numbered freeways, though. I think you misunderstand me. I'm a Borland/Inprise fan from way back. My first C environment, my first Pascal environment, my first Windows development environment, and my first relational database environment all came from Borland. I've a lot of respect for them. I've subsequently switched to MS Visual C on Windows (and gcc everywhere else) because Borland^H^H^H^H^H^H^H Inprise got too caught up in the "radpid development" craze to release a decent C compiler for Windows that could keep up with Platform SDK changes. That said, it is excusable for neither Sun nor Borland nor Microsoft to stamp "100% Pure Java -- Write once, Run anywhere" on an application that only runs on three very specific hardware/software combinations. Wasn't the whole point of Java platform independence? Assuming it is 100% Java, shipping the app as a JAR file would solve this. I'm calling their bluff, which is why I'm bitching. This is especially bad form for Borland, who had a pretty decent Windows->OS/2 PM cross-compiler a few years back. It would actuall translate Windows API code into the appropriate OS/2 PM calls. Cross-platform used to be a breeze for them to do very well.
fully functional Java IDE. It's available for download now at the Inprise Web site. This version is 100% Java
Except for the installation kit, which bombed on my machine
and runs on Linux,
Only if you're an x86-user running RedHat (eck, yech, blechhh) and have KDE installed. If you run an Alpha, MIPS, or S/Linux box, you're S.O.L.
Solaris
but only on UltraSPARC hardware
and Windows.
Well, Windows NT, anyway.
So... let's see. If I'm running an Intel chip, I can choose between Windows NT (slow) and RedHat Linux (why? When there's so much better to be had?) with KDE installed. If I've got a SPARC-based computer, I can only use it on Solaris.
Yay! Let's hear it for "write once, run anywhere!". Distributing a binary installation kit, I can understand (for newbies), but what would be wrong with a huge-ass JAR file and a tarball of HTML documentation? I mean, if you strip away all the installation kits and things, it should run on my platform, right? I mean, it is 100% Java, right?
I think I'll stick with Blackdown on my S/Linux box for now.
What's far more depressing is that it's not even hand coded or running on a decent server. Adobe Pagemill generated the page, and Microsoft IIS served it up. That has to be the most expensive chunk of HTML that I've ever seen.:) I have my doubts about the technical (also ethical) prowess of a government security agency that runs an insecure server OS and can't hand-code a "test page". Perhaps they're just a bunch of 5kR!P7 K1Dd13Z? At least I'm safe.... the machine I connected with was running Win98, and can't possibly stay up for long enough between reboots for them to do anything.:)
An easier way to take care of this would be to tell your DNS server that it is authoritative in the etoys.com domain. Then, pull a copy of the records from some site like kinky-pron.com, and tell your DNS server (easy if you're running BIND) that this file contains the records for etoys.com. I mean, an unfortunate mistake like that (maybe you added a porn site to your service in the domain "etois.com" and misspelled it completely by accident). Now, if you're a sizable ISP, you've directed a large number of people to a porn site in "etoys.com" how many of those people (assuming they want toys for their children) will return? Remember the Zelda.com fuss when Zelda 64 came out? Things could get "interesting". And, no, I am not speaking on behalf of my employer in this post.:)
As far as I know, the filesystem-inspecific optimization routines present in Windows NT had been written by Executive Software back for NT 3.51 Service Pack 2. From what I remember, to run Diskeeper on Windows NT 3.51 (or was it 3.5?), you needed to install an Executive-software specific kernel to replace the Microsoft-standard Windows NT kernel. The Executive Software kernel added a few NT-native function calls that talked directly to the filesystem drivers to move blocks of data around and return various statistics (the sort of which generally aren't of interest to software running atop to OS, but vital to the OS itself). Basically, it was a layer of abstraction that let priviledged processes talk to the FS drivers through a standard interface (sort-of like ioctl()IIRC, Microsoft thought that this was a good enough idea (so that someone else could also write a defragmenter for NT) to include in the Microsoft kernel in a certain service release of NT 3.51 (I think it was either SP2 or SP3). Those functions were carried over into Windows NT 4.0. So basically, all versions of Windows NT since 3.51 SPwhatever have had some code written by Executive, albeit not as much as a whole defragmenter. Why now is the German government upset? Is there a lines-of-code limit to how much Scientolgist-written code may exist within a product?
I think Corel is just trying to protect their ass.
A noble endeavour indeed, as it is currently protecting their head.
Think about it, when you where 13 and got a fresh copy of Debain 1.3 or older slackware, you where excited right? Then what happened? Yes I did the same thing as many of you, I was so excited I installed it
with out reading the instructions and hosed the other OS on the computer I was installing it on.
You couldn't read at 13? And, even if you did nuke a machine, I'm sure you learned:
A decent bit about how an OS and computer interact with each other
The peace-of-mind gained through recent backups
To always read the instructions on software installations.
What if that other OS was your parents workstation and had years of xyz data that they used for work? It would be gone, but you know what, Corel would be able to say, "Hey it illegal downloaded that software from our ftp site, it is not all fault he hosed the install, it is his"
And the last time you heard of a software vendor taking responsibility for anything their software did was..... when? Look at the last paragraph of almost any EULA and find out that the "(vendor) disclaims all warranties, whether implied or expressed, including merchantability or fitness for any purpose". Microsoft, Inprise, and a thousand other vendors claim no warranty at all without placing any explicit age limitations in their EULAs.
I know alot of people, including my self, that at a young age tried to quickly install Linux on there parents computers without taking the time to read all the instructions (and warning messages) on fips and fdisk.
And I know a good many (including myself) who experimented, at a young age, with Linux-based operating environments with very favourable results. Age doesn't innately breed intellect, patience, or wisdom. As someone who has worked the technical support desk when the reps are busy (I do network administration now), I can assure you that age has nothing to do with ignoring error messages, ignoring instructions, and completely trashing PCs. More than anything, it's experience. How are people to gain experience if liceneses start to prohibit them from using software that makes such an excellent learning tool?
Corel is a company in the Business world and they don't want to be held responsiable for there product, like so many other companies.
I really hate that term: "business world". Really.... what does it mean other than "we screw the little guy and pass the savings on to you!"? That sort of justification is precisely what the GPL has been developed to combat. Corel is missing the point of GPLed software, and the "spirit of Linux" in general. *Sigh* Is this really better than when Linux was just a "hacker's" OS? Corporate penetration and all, what has it gained us other than favourable words from ZD-Net and a bunch of suits asking for tech support from RedHat? From a Sun Microsystems bug report (#4102680):
Me, Too! From a Sun Microsystems bug report (#4102680):
Re:Flavours of BSD, Why so many?
on
Which BSD?
·
· Score: 1
This would be largely true if, in fact, all the free BSD implementations were, in fact, coordinating forks, with an open, common parent. This isn't the case. These aren't "distributions" or "versions" of some common BSD set of tools. They are largely complete reimplementations of BSD 4.2 or BSD 4.4. Each saw different, largely orthogonal, shortcomings in the BSD "stardard" package, and decided to release a reimplementation that addressed those issues. It wasn't decided from on-high that "Now there shall be OpenBSD for the security-conscious, FreeBSD for the Intel users, and NetBSD for people with esoteric hardware." Each group decided to form themselves. That said, from what I've heard, there is a decent bit of cooperation (or even mimicking) amongst the groups when some best-thing-since-sliced-bread improvement is made. BTW, "regular" BSD (I'm assuming you mean BSD/OS) is so far behind the times that its laughable. Some of it's new features in 4.0 include ELF support, GCC 2, and support for the 3Com 509 and SoundBlaster 16. From a Sun Microsystems bug report (#4102680):
No, actually Slashdot admins and readers hate www.polyester.net because it spam-posts on completely unrelated Slashdot stories. You do realize that we could, as a collective, do the following:
But, we're a very civilized group and would never think of getting you in trouble with your ISP, right? Heh.... heh... heh..
Oh well, here's my contribution:
---==== Begin Silly-Taunt ====---
So, go and boil your bottoms, sons of a silly person! I waive my genitals at your aunties! Your mother was a hamster, and your father smelt of elderberries.
Now, go away or I shall taunt you a second time-a.
That was, indeed, a refreshing reprive from the flame-fodder that usually pollutes this site. You speak with both eloquence and truth, and I lament only not having said these things before you. Thank you, sir.
Okaaaay... Perhaps I'm missing something here, but just exactly why did this make Slashdot's "news-worthy" cut?
Maybe the link's wrong, or it's written in a languagy syntactically identical to English where all the words have different menaing, or something because all it looked like to me was a lamer suit-type whining about his latest conspiracy theory.
Case in point: Our friend the author here seems to think that since HotMail (TM and (R) as necessary) is an Internet-based service, it is inherently less secure than PC-based email. Okay, here's a question. Before I click that "Check for new mail" widget, where is my mail? OH MY GOSH! It's out there on that scary Internet! ARRRGH!
Okay, that sort-of nullifies his whole argument. Email is spooled on networked machines anyway, not sent directly from workstation to workstation. He fails to realize that all email has the same potential risk, and the first line-of-defense has much to do w/ quality of server software, and network security. These things can be fixed to a large extent.
Also, our friend the authordroid seems to be mistaking storing applications on a remote sever with storing data on a remote server. Is there really any problem with accessing an application via network that updates itself automagically and lets you save your data either on the server or locally?
Perhaps, though... the application is really being controlled by pinkos hiding out at Sun who are reading your steamy letters to your girlfriend! Please! Enough with the conspiracy theories! Sun makes workstations (You know, like PCs, only bigger) and operating systems, too. Sun couldn't have possibly purchased Star Division to make StarOffice work better with these products, could they?
No, one shouldn't have to be an auto-technician to drive a car, but you should at least know enough so that you're not completely stranded when your tire blows out, or know who goes first at a four-way stop. Does anyone know how we got to live in a society where people pride themselves on not having to know things?
Since release 5.0 or thereabouts, I'd been calling Red Hat "Sh*t-in-a-box". Maybe if they'd release a product that took up less space than NT and actually ran (without crashing... not "ran" in the Micro$oft sense) on all of their "supported" platforms (ahem, SPARC), I'd think of something kinder to call them.
No, this isn't FUD. I've had problems with RH Linux on SPARC to an insane extent (try running it headless with the default install and see what happens!). At least the compilers they shipped in 6.0 didn't spew crap all over the place when I asked them to compile KDE (some horrid runtime library problems in 5.2).
No thanks, I'll stick with Slackware (or Sharcware, when it's finished) for my Linux experience. Patrick's done a wonderful job of fixing most people's gripes since 3.0 (like that whole thing about the installer ignoring what packages you chose to install/not-install).
And, yes, I've used RH on it's old stomping-grounds (Intel), too. I'm not very impressed. Seems to act too much like Microsoth Windows for Unix. The installer has that black-box mentality, and the overall distribution presents this "I know how I'm supposed to work better than you do" image.
Maybe I should jump the fence and run back to BSD (or SunOS 4;) ).
Just for the hilarity factor! I agree with the above poster's sentiments (C for CGI and Java for the Rest), but he expressed it much more.... succintly than I ever could.:)
And, yes, I have used PERL and made some very functional works out of it--the language simply doesn't fit my mindset. I just genuinely like C better for CGIs (execution time, anyone?) and Java for things that need much user-interaction. PERL just seems to have all the code-beauty of AWK with all the functionality of BASIC (including the exponential code-bloat).
I mean, come on, if you must use an interpreted language, at least use LISP or Guile, something that has a little more style to it and isn't a write-only language.
Yeah, but those pesky CTX monitors are much harder to fold up with your laptop than a 15" LCD display panel is. The panel also has the benefit of being made into the laptop, so you don't need an extra power cable.
Uhm.. I think you meant to post to segfault.org, not slashdot.org. The lack of Natalie Portman's genitals and Hank the Angry, Drunken Dwarf should have been a clue. Jeeez....
Out-of-Box, Suns don't make nearly as nice workstations as SGIs do (I've used both). But, with a little tweaking, they can make a wonderful home. I'm on a SPARCplug (okay, it's not a Sun, but it's based on Sun's technology and runs their software and pretends it's an SS20) running Solaris 7, KDE, and Netscape/IE (depending on my mood). I really couldn't be much happier.
Ultras suck as workstations when they only have the standard TGX framebuffer (stick an old Paradise ISA VGA card in a PentiumII and see if it's really that much more responsive in Windows than a P60 is... you catch my drift). TGXs tend to bug-out on their onboard RAM (mine went out a month ago, so now I'm at half RAM, which means -less- acceleration), which produces very `interesting' video effects (mouse droppings, incomplete redraws, region-swapping, etc.).
However, throw a Creator3D or Elite3D in the Ultra, and you're moving like nobody's business. Suns really shine when you run them headless and X into them or run remote processes on them, as their I/O is superb and their CPUs run phenomenally once you free them from the tethers of a video framebuffer.
The keyboards are weird, especially if it's the Unix layout (and you're a PC-user), but, once you acclimate to them (Sun makes a PC101-like keyboard, BTW), they're a dream in Solaris. I'd love to have copy/paste/help/undo/redo/repeat/open/cut/properti es keys on a WinNT box... it saves from having to remember shortcut keys. But, the keyboards are an acquired taste
As for SGIs, they do make much more impressive workstations. They've got glamourous cases, the spiffiest GUI on the planet, a standard PC keyboard, and a graphics subsystem that works with the CPU (actually uses the same instruction set), instead of adding more work to the CPU. Excellent, excellent workstations.
But, run IRIX machines aren't nearly as impressive remotely or as servers. Challenges and the like were pretty darn good, but, you get more bang-for-buck on Suns if you're doing mostly serving or headless processing.
People buy Suns for CPU speed and Solaris, not for a spiffy GUI or `creature comforts' (well, most don't... I actually enjoy working in the Solaris environment). People buy SGIs because they need the power and flexibility of a Unix workstation, but also want a machine that makes computing an far more enjoyable experience (in terms of `creature comforts').
Oh.. don't talk about stuffing PC innards into an SGI case too loudly around avid SGI users (I'm one of those, too). They just might burn you at the stake for heresy.;)
The Sun Power Key (actually suspend/power) is above the keypad numeric keypad on the sun5 keyboards (next to contrast, volume, and monitor-on keys). It turns on the newer systems (with soft power support), and enters hibernate (save memory state and shut down) mode on machines with APM capability. You can disable its suspend function (trust me... bumping that when reaching for your Coke is very annoying, since it doesn't confirm shutdown) by simply uninstalling the driver for it.
The Power Key has the universal interrupted-circle (or intersecting 0/1, depending on your point-of-view) symbol on it. I think we're talking about the `mystery key' above the escape key. It's the `null key'... some Sun hacker's idea of a useful joke (a possible allusion empty lists in LISP... they're lists, but lists of nothing). It doesn't do anything most of the time, but returns an ASCII nul (0x00) character if your program asks nicely for it.
Slashdot Mirror
I recall a couple of years back, surrounding the release of Windows 95, an interview with Bill Gates on ABC's 20/20. The end of the interview closed with the interviewer being practically pushed out of Bill's office with Bill screaming at his secretary/assistant "I said five minutes. He had his five minutes, now he has to go. Mister? I've given you five minutes." It sounded a lot like something a young child throwing a temper tantrum after he'd already been cleaning his room "Forever". It makes you wonder if the Microsoft corporate hierarchy doesn't consist of people yelling downwards at the serfs who churn out thousands of lines of code and tens of pages of documentation a day.
How many times have you heard a Macintosh zealot give Windows users a hard time because nearly every Windows user-interface feature appeared in MacOS first? It is just me, or does the new Finder look a lot like the hideously-slow Microsoft "Active Desktop"? Everything's contained in one window (that looks strangely like a web browser). Big, hideous buttons/icons for everything. Lots of extraneous eye candy. As an ex-Mac user, I always loved the way Apple kept their UI clean. Things were kept simple (IE: fast and well thought-out) yet elegant. Even the idea of "exploding" windows was a great UI clue to the user to answer "What made that pop up?". But this? Have they any idea how many long-time Mac users they will alienate by tearing up the user interface that was even present on the old Lisa and Apple (not Mac) IIgs? The site mentions extending the "one window" metaphor to the entire operating system because "you can only interact with one application at a time". Doesn't this sound a lot like the days before the Multifinder? Wasn't one of System 7's greatest achievements the ability to multitask visually? The idea of BSD running behind it all sounds excellent, as Apple's OS has had a lot of holes in its design (seeming inability to use the hardware MMUs correctly, for one thing). However, they've taken what seems to be an obvious leap backwards in UI design by eliminating the clarity of pupose that every visible item in the Mac UI used to have. As a user, and as someone who has to provide support for users, I certainly hope there's an "Old Finder Look-and-Feel" option in "General Controls". I mean, come on. Apple's incarnation of the "Desktop" metaphor has propogated to OS/2, Windows, and who-knows-how-many window managers. Isn't tossing all this out the window just "thinking" a bit "too different"?
Dr.Scheme, Rice University's implementation of MzScheme (also known as Rice Scheme) is now up to version 101. Back when I took COMP 212 there (the class that uses Dr.Scheme exclusively), the software was at version 53. That was only Fall 1998, so perhaps they've confused "build number" with "version number". :)
- The annoying bastards continue until the signal-to-noise ratio here is unbearable (we only have a finite number of moderator points)
- People turn up their thresholds to the point that if you're not Linus Torvalds, Bruce Perens, or RMS, your post won't even get read.
- We continue to bitch about the idiots
Of course, another obvious answer would be to implement a "cancel post" feature accessible to a few super-moderators that would effectively "kill" a post by setting its score to -2 or -infinity. That's an awful lot of power, though, and I think it would sit horribly with the Slashdot community at large. I know this is long and somewhat off-topic (although this article beautifully demonstrates the need for action), but I think it's important. Please post your ideas as a follow-up. It's time to take our forum back! PS: If someone sees this post as worthwhile and knows of a better place to post/send it, please let me know. Do the obvious thing to my posted email address to contact me. Spam and flamage not welcome, but coherent comments always are.I have to admit that the start of my hate of RedHat comes from my experiences with it on the SPARC platform. RedHat 5.something (not their first release on SPARC, but their first in quite a while) shipped so broken out-of-box that I ended up having to cross-compile half of the libraries on a Solaris machine just so that I could reliably compile. Eventually, I got the machine running (keep in mind... I'm not a newbie. I've been an administrator on Solaris, SunOS 4, IRIX 5, IRIX 6, and Linux for several years), and I've tried to poke the configuration as little as possible. Hopefully, over this Christmas holiday, I'll have time to put either Debian or Solaris 7 on the machine. As far as other distributions go, I'm a big fan of Slackware on the x86 platforms. Installation's a snap, it installs only what you ask for. I've walked people through installations of RedHat where they specifically turned off any X support, and RH decided that some subcomponent of some package needed X for a frontend. So, RH installs X, and X needs this and that and something else, and fonts, and blah-blah-blah. I've briefly played with Debian on the SPARC. Its installation routine is definitely easier than RedHat's, although using dselect later to add packages can be confusing if you don't read the README files first. There's also the lack of precompiled software for Debian, as RedHat's RPM is very popular. Although, being a SPARC user, I can rarely find S/Linux RPMs, anyway. I guess my main gripe with RedHat is that they're showing their "true self" as a traditional American corporation, instead of an organization dedicated to furthering the sensible development of an operating environment. They tend to suffer from shovelware syndrome, have a tendency to ship OSes with beta versions of the C runtime library, and have very-much a black box mentality when customizing the OS. This all well-and-good for a new user (or is it?) who doesn't know his way around the OS, but it's damned frustrating to a Unix veteran. I'd been trying to port Slackware to the SPARC platform. However, Patrick Volkerding wasn't interested in blessing my work, as SuSE had burned him before in such a deal. That, combined with my workload caused me to all but abandon the probject. I dunno... people bitch about Slackware because it doesn't behave like a System V Unix. I bitch about Redhat because it behaves like "RedHat OS" instead of like a Unix-like operating environment. I guess it all boils down to the fact that RH 5.0 was utter crap, and they lost me as a customer because of it's instability on both SPARC and Intel platforms. They have this get-out-the-door-and-ship-erratas-later mentatlity that has all but killed HP-UX's credibility and continues to plague Microsoft. I tend to like things that work well out-of-box. If I have to recompile sendmail or ftpd because of some new exploit, that's perfectly acceptable. However, if I can't compile out of box because libc.a or the compiler is broken, that's simply not acceptable, IMHO.
Actually, I'm complaining about the fact that a supposedly "pure-Java" application is platform-specific. Isn't this what landed Microsoft in the fire a few months back? If you gave a car, I wouldn't bitch at the colour. I'd bitch if it only took left-exits onto prime-numbered freeways, though. I think you misunderstand me. I'm a Borland/Inprise fan from way back. My first C environment, my first Pascal environment, my first Windows development environment, and my first relational database environment all came from Borland. I've a lot of respect for them. I've subsequently switched to MS Visual C on Windows (and gcc everywhere else) because Borland^H^H^H^H^H^H^H Inprise got too caught up in the "radpid development" craze to release a decent C compiler for Windows that could keep up with Platform SDK changes. That said, it is excusable for neither Sun nor Borland nor Microsoft to stamp "100% Pure Java -- Write once, Run anywhere" on an application that only runs on three very specific hardware/software combinations. Wasn't the whole point of Java platform independence? Assuming it is 100% Java, shipping the app as a JAR file would solve this. I'm calling their bluff, which is why I'm bitching. This is especially bad form for Borland, who had a pretty decent Windows->OS/2 PM cross-compiler a few years back. It would actuall translate Windows API code into the appropriate OS/2 PM calls. Cross-platform used to be a breeze for them to do very well.
as in "free beer"
Except for the installation kit, which bombed on my machine
Only if you're an x86-user running RedHat (eck, yech, blechhh) and have KDE installed. If you run an Alpha, MIPS, or S/Linux box, you're S.O.L.
but only on UltraSPARC hardware
Well, Windows NT, anyway.
So... let's see. If I'm running an Intel chip, I can choose between Windows NT (slow) and RedHat Linux (why? When there's so much better to be had?) with KDE installed. If I've got a SPARC-based computer, I can only use it on Solaris.
Yay! Let's hear it for "write once, run anywhere!". Distributing a binary installation kit, I can understand (for newbies), but what would be wrong with a huge-ass JAR file and a tarball of HTML documentation? I mean, if you strip away all the installation kits and things, it should run on my platform, right? I mean, it is 100% Java, right?
I think I'll stick with Blackdown on my S/Linux box for now.
What's far more depressing is that it's not even hand coded or running on a decent server. Adobe Pagemill generated the page, and Microsoft IIS served it up. That has to be the most expensive chunk of HTML that I've ever seen. :) I have my doubts about the technical (also ethical) prowess of a government security agency that runs an insecure server OS and can't hand-code a "test page". Perhaps they're just a bunch of 5kR!P7 K1Dd13Z? At least I'm safe.... the machine I connected with was running Win98, and can't possibly stay up for long enough between reboots for them to do anything. :)
An easier way to take care of this would be to tell your DNS server that it is authoritative in the etoys.com domain. Then, pull a copy of the records from some site like kinky-pron.com, and tell your DNS server (easy if you're running BIND) that this file contains the records for etoys.com. I mean, an unfortunate mistake like that (maybe you added a porn site to your service in the domain "etois.com" and misspelled it completely by accident). Now, if you're a sizable ISP, you've directed a large number of people to a porn site in "etoys.com" how many of those people (assuming they want toys for their children) will return? Remember the Zelda.com fuss when Zelda 64 came out? Things could get "interesting". And, no, I am not speaking on behalf of my employer in this post. :)
As far as I know, the filesystem-inspecific optimization routines present in Windows NT had been written by Executive Software back for NT 3.51 Service Pack 2. From what I remember, to run Diskeeper on Windows NT 3.51 (or was it 3.5?), you needed to install an Executive-software specific kernel to replace the Microsoft-standard Windows NT kernel. The Executive Software kernel added a few NT-native function calls that talked directly to the filesystem drivers to move blocks of data around and return various statistics (the sort of which generally aren't of interest to software running atop to OS, but vital to the OS itself). Basically, it was a layer of abstraction that let priviledged processes talk to the FS drivers through a standard interface (sort-of like ioctl()IIRC, Microsoft thought that this was a good enough idea (so that someone else could also write a defragmenter for NT) to include in the Microsoft kernel in a certain service release of NT 3.51 (I think it was either SP2 or SP3). Those functions were carried over into Windows NT 4.0. So basically, all versions of Windows NT since 3.51 SPwhatever have had some code written by Executive, albeit not as much as a whole defragmenter. Why now is the German government upset? Is there a lines-of-code limit to how much Scientolgist-written code may exist within a product?
- A decent bit about how an OS and computer interact with each other
- The peace-of-mind gained through recent backups
- To always read the instructions on software installations.
And the last time you heard of a software vendor taking responsibility for anything their software did was..... when? Look at the last paragraph of almost any EULA and find out that the "(vendor) disclaims all warranties, whether implied or expressed, including merchantability or fitness for any purpose". Microsoft, Inprise, and a thousand other vendors claim no warranty at all without placing any explicit age limitations in their EULAs. And I know a good many (including myself) who experimented, at a young age, with Linux-based operating environments with very favourable results. Age doesn't innately breed intellect, patience, or wisdom. As someone who has worked the technical support desk when the reps are busy (I do network administration now), I can assure you that age has nothing to do with ignoring error messages, ignoring instructions, and completely trashing PCs. More than anything, it's experience. How are people to gain experience if liceneses start to prohibit them from using software that makes such an excellent learning tool? I really hate that term: "business world". Really.... what does it mean other than "we screw the little guy and pass the savings on to you!"? That sort of justification is precisely what the GPL has been developed to combat. Corel is missing the point of GPLed software, and the "spirit of Linux" in general. *Sigh* Is this really better than when Linux was just a "hacker's" OS? Corporate penetration and all, what has it gained us other than favourable words from ZD-Net and a bunch of suits asking for tech support from RedHat?From a Sun Microsystems bug report (#4102680):
Me, Too!
From a Sun Microsystems bug report (#4102680):
This would be largely true if, in fact, all the free BSD implementations were, in fact, coordinating forks, with an open, common parent. This isn't the case. These aren't "distributions" or "versions" of some common BSD set of tools. They are largely complete reimplementations of BSD 4.2 or BSD 4.4. Each saw different, largely orthogonal, shortcomings in the BSD "stardard" package, and decided to release a reimplementation that addressed those issues. It wasn't decided from on-high that "Now there shall be OpenBSD for the security-conscious, FreeBSD for the Intel users, and NetBSD for people with esoteric hardware." Each group decided to form themselves. That said, from what I've heard, there is a decent bit of cooperation (or even mimicking) amongst the groups when some best-thing-since-sliced-bread improvement is made. BTW, "regular" BSD (I'm assuming you mean BSD/OS) is so far behind the times that its laughable. Some of it's new features in 4.0 include ELF support, GCC 2, and support for the 3Com 509 and SoundBlaster 16.
From a Sun Microsystems bug report (#4102680):
No, actually Slashdot admins and readers hate www.polyester.net because it spam-posts on completely unrelated Slashdot stories. You do realize that we could, as a collective, do the following:
But, we're a very civilized group and would never think of getting you in trouble with your ISP, right? Heh.... heh... heh..
Oh well, here's my contribution:
---==== Begin Silly-Taunt ====---
So, go and boil your bottoms, sons of a silly person! I waive my genitals at your aunties! Your mother was a hamster, and your father smelt of elderberries.
Now, go away or I shall taunt you a second time-a.
From a Sun Microsystems bug report (#4102680):
If, by telling the trolls to go elsewhere, I am branded a troll, does this not also brand BIG> Moderators as trolls?
Just a thought.
From a Sun Microsystems bug report (#4102680):
That was, indeed, a refreshing reprive from the flame-fodder that usually pollutes this site. You speak with both eloquence and truth, and I lament only not having said these things before you. Thank you, sir.
From a Sun Microsystems bug report (#4102680):
Have you considered a career as a suicide victim? Or, at least considered taking your worthless banter to a more appropriate place?
From a Sun Microsystems bug report (#4102680):
Okaaaay... Perhaps I'm missing something here, but just exactly why did this make Slashdot's "news-worthy" cut?
Maybe the link's wrong, or it's written in a languagy syntactically identical to English where all the words have different menaing, or something because all it looked like to me was a lamer suit-type whining about his latest conspiracy theory.
Case in point: Our friend the author here seems to think that since HotMail (TM and (R) as necessary) is an Internet-based service, it is inherently less secure than PC-based email. Okay, here's a question. Before I click that "Check for new mail" widget, where is my mail? OH MY GOSH! It's out there on that scary Internet! ARRRGH!
Okay, that sort-of nullifies his whole argument. Email is spooled on networked machines anyway, not sent directly from workstation to workstation. He fails to realize that all email has the same potential risk, and the first line-of-defense has much to do w/ quality of server software, and network security. These things can be fixed to a large extent.
Also, our friend the authordroid seems to be mistaking storing applications on a remote sever with storing data on a remote server. Is there really any problem with accessing an application via network that updates itself automagically and lets you save your data either on the server or locally?
Perhaps, though... the application is really being controlled by pinkos hiding out at Sun who are reading your steamy letters to your girlfriend! Please! Enough with the conspiracy theories! Sun makes workstations (You know, like PCs, only bigger) and operating systems, too. Sun couldn't have possibly purchased Star Division to make StarOffice work better with these products, could they?
No, one shouldn't have to be an auto-technician to drive a car, but you should at least know enough so that you're not completely stranded when your tire blows out, or know who goes first at a four-way stop. Does anyone know how we got to live in a society where people pride themselves on not having to know things?
By the way, Mr. Stalder, that's HotMail Crack.
From a Sun Microsystems bug report (#4102680):
Since release 5.0 or thereabouts, I'd been calling Red Hat "Sh*t-in-a-box". Maybe if they'd release a product that took up less space than NT and actually ran (without crashing... not "ran" in the Micro$oft sense) on all of their "supported" platforms (ahem, SPARC), I'd think of something kinder to call them.
No, this isn't FUD. I've had problems with RH Linux on SPARC to an insane extent (try running it headless with the default install and see what happens!). At least the compilers they shipped in 6.0 didn't spew crap all over the place when I asked them to compile KDE (some horrid runtime library problems in 5.2).
No thanks, I'll stick with Slackware (or Sharcware, when it's finished) for my Linux experience. Patrick's done a wonderful job of fixing most people's gripes since 3.0 (like that whole thing about the installer ignoring what packages you chose to install/not-install).
And, yes, I've used RH on it's old stomping-grounds (Intel), too. I'm not very impressed. Seems to act too much like Microsoth Windows for Unix. The installer has that black-box mentality, and the overall distribution presents this "I know how I'm supposed to work better than you do" image.
Maybe I should jump the fence and run back to BSD (or SunOS 4 ;) ).
From a Sun Microsystems bug report (#4102680):
Just for the hilarity factor! I agree with the above poster's sentiments (C for CGI and Java for the Rest), but he expressed it much more.... succintly than I ever could. :)
And, yes, I have used PERL and made some very functional works out of it--the language simply doesn't fit my mindset. I just genuinely like C better for CGIs (execution time, anyone?) and Java for things that need much user-interaction. PERL just seems to have all the code-beauty of AWK with all the functionality of BASIC (including the exponential code-bloat).
I mean, come on, if you must use an interpreted language, at least use LISP or Guile, something that has a little more style to it and isn't a write-only language.
"Have lunch, or be lunch."
Yeah, but those pesky CTX monitors are much harder to fold up with your laptop than a 15" LCD display panel is. The panel also has the benefit of being made into the laptop, so you don't need an extra power cable.
"Have lunch, or be lunch."
Uhm.. I think you meant to post to segfault.org, not slashdot.org. The lack of Natalie Portman's genitals and Hank the Angry, Drunken Dwarf should have been a clue. Jeeez....
"Have lunch, or be lunch."
Out-of-Box, Suns don't make nearly as nice workstations as SGIs do (I've used both). But, with a little tweaking, they can make a wonderful home. I'm on a SPARCplug (okay, it's not a Sun, but it's based on Sun's technology and runs their software and pretends it's an SS20) running Solaris 7, KDE, and Netscape/IE (depending on my mood). I really couldn't be much happier.
Ultras suck as workstations when they only have the standard TGX framebuffer (stick an old Paradise ISA VGA card in a PentiumII and see if it's really that much more responsive in Windows than a P60 is... you catch my drift). TGXs tend to bug-out on their onboard RAM (mine went out a month ago, so now I'm at half RAM, which means -less- acceleration), which produces very `interesting' video effects (mouse droppings, incomplete redraws, region-swapping, etc.).
However, throw a Creator3D or Elite3D in the Ultra, and you're moving like nobody's business. Suns really shine when you run them headless and X into them or run remote processes on them, as their I/O is superb and their CPUs run phenomenally once you free them from the tethers of a video framebuffer.
The keyboards are weird, especially if it's the Unix layout (and you're a PC-user), but, once you acclimate to them (Sun makes a PC101-like keyboard, BTW), they're a dream in Solaris. I'd love to have copy/paste/help/undo/redo/repeat/open/cut/properti es keys on a WinNT box... it saves from having to remember shortcut keys. But, the keyboards are an acquired taste
As for SGIs, they do make much more impressive workstations. They've got glamourous cases, the spiffiest GUI on the planet, a standard PC keyboard, and a graphics subsystem that works with the CPU (actually uses the same instruction set), instead of adding more work to the CPU. Excellent, excellent workstations.
But, run IRIX machines aren't nearly as impressive remotely or as servers. Challenges and the like were pretty darn good, but, you get more bang-for-buck on Suns if you're doing mostly serving or headless processing.
People buy Suns for CPU speed and Solaris, not for a spiffy GUI or `creature comforts' (well, most don't... I actually enjoy working in the Solaris environment). People buy SGIs because they need the power and flexibility of a Unix workstation, but also want a machine that makes computing an far more enjoyable experience (in terms of `creature comforts').
Oh.. don't talk about stuffing PC innards into an SGI case too loudly around avid SGI users (I'm one of those, too). They just might burn you at the stake for heresy. ;)
"Have lunch, or be lunch."
The Sun Power Key (actually suspend/power) is above the keypad numeric keypad on the sun5 keyboards (next to contrast, volume, and monitor-on keys). It turns on the newer systems (with soft power support), and enters hibernate (save memory state and shut down) mode on machines with APM capability. You can disable its suspend function (trust me... bumping that when reaching for your Coke is very annoying, since it doesn't confirm shutdown) by simply uninstalling the driver for it.
The Power Key has the universal interrupted-circle (or intersecting 0/1, depending on your point-of-view) symbol on it. I think we're talking about the `mystery key' above the escape key. It's the `null key'... some Sun hacker's idea of a useful joke (a possible allusion empty lists in LISP... they're lists, but lists of nothing). It doesn't do anything most of the time, but returns an ASCII nul (0x00) character if your program asks nicely for it.
"Have lunch, or be lunch."