OK, the major problem I see is this... if you are running an old 14,000 bps modem, then maybe... but
otherwise, the manufacturers would have to have used LEDs that support on-off-on 100 million times a second (plus of course all the other data that encapsulates the data itself (like the bits that are part of the TCP/IP packet and not the data itself)...
Regular LED's of course are designed to do this... oh, wait, they arent!
Wow! First, thanks for the first legitimate response to my post!
As every decent programmer should know, many (most) programs call API's that are part of the OS to do their work on virtually any GUI based OS. Windows is no different. That IE calls API's that other Windows code and other apps use is irrelevant to whether it is "integrated" into the OS. Sharing or using the OS's API's is not "integrated" otherwise almost every program installed on a computer could be called "integrated"
I guess it depends on your definition of "operating system". Almost any part of the operating system can be replaced, but at what cost? A new virtual memory manager could be swapped in, but what if that entailed kernel changes that would affect other parts of the OS? Would you then consider the VMM to be fully integrated or simply another replaceable part?
Now I'm not saying that IE goes down to that low of a level, but in a way they have integrated it into the higher level of the OS -- as a part of the various controls you can build into your app. IE itself is just another application running on your machine, but it uses the core controls that are built into the OS. As far as I'm aware, removing those controls would be like saying you no longer want radio buttons or tree-view controls. Can it be done? Yes, but at a potentially large cost to the user and development community.
>Trueexcept that Windows had those controls
>since before IE existed. It also had a File Mangler
Additionally, numerous companies and the DOJ have already proven IE can be de-"integrated" from the OS,. resulting in a speed increase, contrary to the falsified MS video to prove otherwise.
I really question this sentiment. How can you prove that something can be completely removed from the OS without modifying the source code? Are these people just proving that the IE app can be removed or that the actual browser control functionality can be completely stripped out?
>APIs that are shared by the OS dont need to
>be removed. The core APP (iexplore.exe, etc)
>and APIs in non core DLLs can be removed.
>Or, as I suggested in another post, even allowing
>the choice of a default browser that the
IE WinGUI interfaceopens for URL requests
>should even satisfy such a requirement.
If they're acutally completely stripping core browser functionality, have these people proven that their actions don't effect that thousands of other Windows apps out there?
Nor is there a reason why MS couldnt write a decent file manager that doesnt need IE - or also as suggested here, at least while using IE as a file manager
See, now you're getting into territory that's wholly debateable. Is it right to ask MS to rewrite parts of the OS at a low level that may take advantage of the functionality they've built into their common controls? Maybe the file manger CAN be rewritten, but just because it can doesn't mean MS should be forced into such a task. Depending on how deep the tendrils go, redoing the file manager may be a much more dauting task than we might think.
>I think, based off the court case where that
>integration was already proven to be for
anti-competitive
>reasons and deemed illegal, that it would be
>a fitting penalty - besides, there are enough
>shareware file browsers out there to prove the
>job isnt that difficult - unless MS programmers
>are really unqualified in their jobs.
I dunno, I've never studied Inside Windows 2000 or other books to see how far this stuff goes.
Right or wrong, MS has managed to get the functionality integrated with other common controls. Just pop open Visual Studio and see how easy it is to get an MFC (blech) app up and running with full browser integration. And when you take a peek at IEXPLORE.EXE in the Dependency Walker, you'll see most of the same functions that you would be using in an app that brought in a browser control.
I think the issue that MS is trying to bring up is this : IE can't just be deinstalled because much of the functionality has been incorporated into the OS via other channels.
>Or IE could be de-installed, and the API's
>that are shared by other apps could be left. Unless
>the OS is truly so poorly documented and the code
>so poorly maintained, this should be a trivial task to
>separate those APIs into those 2 baskets.
- Robert
IE itself is using that functionality and so can other apps. Removing the parts that other apps use could potentially affect a large number of users and developers. The real thing someone needs to prove is that IE and all of its functionality can be taken out and still have Windows and other apps function like nothing was missing. I personally don't know if that's the case or not.
Like Windows OEM release when IE 4.0 was late for inclusion in the normal installer and was installed (if the OEM decided to, or the end user at a later date) from the second CD that came in the OEM bundle.
We still have customers who are not using the IE interface because they never ended up installing IE in that fashion to "integrate" it.
What's wrong is that isnt what happened - as was evidenced by the various programs that remove IE, and the early Windows pre-IE 4 releases where the IE "integration" was done after the install by using a second CD. No program I know of fails to run on those Win98 systems unless they arent programs but instead scripts or such that use IE as a pretend interface (like many CD catalog browsers and document browsers).
Besides, your claim implies that MS has published the IE APIs for others to use, these APIs are in now in the core API set by your example making removal of the IE executable and the IE API set in it's own DLLs still quite removable.
Unfortunately, I think your post irrelevant at best as it discusses how things could be done, not how it actually happened in the Windows world.
As every decent programmer should know, many (most) programs call API's that are part of the OS to do
their work on virtually any GUI based OS. Windows
is no different. That IE calls API's that other
Windows code and other apps use is irrelevant to
whether it is "integrated" into the OS. Sharing or
using the OS's API's is not "integrated" otherwise almost every program installed on a computer could be called "integrated"
Additionally, numerous companies and the DOJ
have already proven IE can be de-"integrated"
from the OS,. resulting in a speed increase,
contrary to the falsified MS video to prove
otherwise.
Add to that, Windows didnt always come with
IE as the shell - and when IE 4.01/4.02 came
out, the IE "integration" came (for many OEM's
such as ourselves) on a second CD. It was, and
is an addon.
Keep in mind, the rest of Windows components
are installed on a "Check this box to install"
method - there is no reason why IE cannot be
installed or not installed in the same fashion.
Nor is there a reason why MS couldnt write a
decent file manager that doesnt need IE - or
also as suggested here, at least while using
IE as a file manager, block it from pulling
http requests - or better yet, make an option
to set and select Window's default browser
and pass all URL's to it via DDE...
Also, I for one, am happy having an
extensible object oriented (not object based)
GUI that doesnt require IE's overhead just
to open a simple drive folder. MacOS, OS/2,
eComStation and some *nix desktop variants
come to mind in that category.
This dead horse keeps coming up. MS knows
they can easily remove IE, it was already proven
in court, MS had to falsify videotaped
tests to try to prove otherwise, were caught
and admitted it, while the DOJ proved it
could be done, and easily.
This sounds a lot like a bad Keanu Reeves movie called "Chain Reaction" - heck, the "fusion reactor" even looks like a scaled down version of the movie prop.
Power Technology (almost 4 years ago) seems to have think they made sufficient advances to warrant claims of far better and lighter batteries... but according to them, the major battery manufacturers arent interested... and I'd presume it would be a ridiculous obstacle financially to try to compete with the big 4 (Rayovac, Eveready, Duracell and Exide). Based off alkaline technology, the batteries would be cheap, light (based off the new chemical composition), and according to PT, capable of filling in the role of lead-acid - the current industrial, heavy use and abuse battery of choice (ie: large scale use, heavy load draw, frequent charge, discharge, continuous use and charge batteries such as in cars, electric forklifts, RV's, home power storage units, UPS's, etc).
http://www.evworld.com/archives/interviews/balak .h tml
(follow links from article for more info)
IBM announced years ago better battery technology - it appeared in the highest end Thinkpads, and is based off Lithium and other elements (it isnt Lithium-ion), but of course, they are too expensive for say... disposable radio batteries...
So, perhaps it is the economies of it that matter... after all, Lexmark, Epson and HP often sell printers at a loss - why? Because they clean up on the consumables... not that I think it is, because I have no knowledge of such, but why wouldnt the battery industry be any different? The less batteries you sell, the less you make. As new battery innovations come out (which usually are negligble and barely if at all noticeable on the "home front" AA, AAA, C, D sized cells), the prices usually go up as well... so though battery life may be increasing slightly, so is the price - at a slightly greater than inflationary rate.
Dont know, and dont care - it's really irrelevant since the item that the article incorrectly touched on was fuel cells, which ARE available and in use for things such as cell phones and other devices already. Supposedly they are more efficient, last longer, and there are methodws of recharging them (though not yet available)... so the technology does exist, and can be used to replace batteries if only more work were put into it. (currently such fuel cells are disposable instead of refillable).
We are far from moderate to large... but I think vample is still right, our users would go crazy - especially with everyone and their mother having hotmail accounts nowadays. Many of our customers (for fear of us selling their email addresses since everyone else does) give us hotmail addresses to send mail receipts to and email problems via those hotmail accounts, so both the owners of sites we host (and thus email accounts) and our own website users with subscription accounts would go crazy.:-( We actually did consider it though.
As for such actions pissing off MS, that'd undoubtedly be the case. Just about everything seems to if it seems a snub to them. We posted a very big, nasty post complaining about their harrassment over their tactics over pirated licenses - which drove one of our suppliers to bankruptcy via lawsuits and lost customers... the supplier was innocent. MS didnt even offer an apology (Micromatix in MD was the company). And of course, the lost customers and revenue were jsut that. Anyway, shortly after our big post and complaint, our servers were attacked with the obvious intent of hacking into them. Oddly, we logged internal MS IP addresses doing the attacks. Since then, we received on a overly regular basis, notice after notice after poster after poster after letter advising us of how MS is cracking down on piracy and if we dont report everyone and anyone we know who is pirating software we will be held liable. The last one was an offer for an "agree to forget you stole our software" if we turn ourselves in. Going on over 10 mailings in as many months.
Here's the kicker... we dont run Windoze. We run MacOS 7.5, 8 and 9.04, Linux, eComStation and OS/2. We DO own 12 NT4 licenses, and 6 Win98 and 3 Win95 licenses though. We just dont use them (reformatted and re-isntalled a different OS).
I'd shudder to think of what the retribution would be if we were a mid sized to large ISP.
On our networks, logging for almost two dozen domains, the largest source of spam via "Open Relay Mail Servers" is Hotmail. These emails are being sent via other servers, and mass mailed via hotmail servers being used to relay them.
Hotmail's responses to the numerous complaints? "We'll cancel that user's account..." Often though it's not the user at fault, since you dont even need a valid Hotmail address to do this. So, even with notifying them of the real problem (open servers) and showing them headers that confirm it, they do nothing.
Our incoming spam would drop by over 45% if they'd fix it.
- Rob
This used to be done off and on in certain locations when AtHome was providing their cable internet. I am hoping it's ended. It was a nightmare on a webserver end as their proxy servers are definitely not up to par. It's bad enough with dealing with AOL's thousands of proxy trackers... errr... servers that ensure that your (a web hosts') documents are being read and re-read and re-read to update the various servers that may be responding to one users' requests.
I'm an idiot, but you arre too stupid to give a single reason why? How odd.
You'd rather run a domain controller on Linux than Warp? How strange. Which one do you run? could it be the same BIND that is available for Warp? No big deal running the same server daemon since there's both the Linux version and a native up to date port available that run on Warp. Personally, I am happy with the one that comes with Warp - especially since one of our customers runs 5 adult sites so the P90 and P133 doing DNS take a beating every day (at 0-1% CPU - guess maybe they are actually quite bored).
Well, glad you like Linux. So do I. But. (1) should I care who else uses Warp? Isnt that the reason so many Linux users ignored as a valid reason why they should go to Windows? (2) Warp has more than stability on it's side. It can run virtually any Linux program, more Win32 apps than Linux as well (since it not just uses Wine, but adds many more APIs the Wine team hasnt gotten to yet that IBM did years ago, PLUS other APIs by the boatload the Project Odin team has done in the last 3 years.) and it runs them natively, by simply clicking their icon to install or run it - no configuration and installation headaches, no Win install, copy the stuff over, edit the registry crap. Just click "Setup" then when done, click the app's icon - done., (3) it makes a far better server than *ANY* PC OS to date - including Linux, sorry to say. (4) it scales to 64 CPU's per node - what else does? Every other 64 CPU setup I have seen has been clusters of 1-8 CPU per node systems. I'd rather cluster an OS that can do 64 per node. (5) I'm not advocating low end hardware - I'm advocating using whatever hardware you have properly (ie, that includes using it FULLY). At dual CPU 300MHz, my Warp system outperforms XP on 1.1GHz. At 133MHz, my Warp Server outperforms XP on 600MHz (with 3 times the RAM). My point is, I'd rather (whether the hardware is low end, high end, or inbetween), take advantage of the fact that Warp (or even Linux for that matter) will run circles around XP. One web server solution for a client to match an identical setup running Warp by us was installed for them by Microsoft. At less traffic, MS ended up installing 6 dual CPU boxes to minimize the "Server too busy" messages. I was happy with our one server, but eventually made it two. The company? Out of business, and bought out by a research company many states away. They were known as America's Doctor, online Doctor Chat and Referrals.
So, lets see, you say I'm an idiot but have yet to say why, or even how MS is making a huge step in the right direction. Yes, XP is more stable than previous releases... but that promise is a decade old - and so what? Turns out it's seeming less secure. 40+ ports open, a dozen known holes and back doors. So, where's the huge step? Maybe it's the hardware incompatibility (which to me includes crippled drivers - which has been the answer to that problem, so while most things ahve drivers, they arent nearly as capable as the Win9X versions.)
As for my other claims? Which ones? Cite one and how I need to "jump at least into the 90's" about it! Warp runs every service Linux does. It runs every service XP does. It even can run ASP stuff - on Apache 2.x no less. It runs the latest of most every server daemon out there - including Apache, MySQL, DB/2, IRC (and even the related eggdrop scripts), Domino Notes Server and more. Oooh - it doesnt run IIS - the least capable under load web server there is. So - seems my claims are quite current considering, we are (with the exception of MS branded and sold servers) talking current releases of virtually every server daemon out there (and that's not including running Linux stuff on Warp). And doing that on a current release of Warp (only a few weeks old).
Nothing is being done. Take a mansion, set the whole thing ablaze and throw a cup of water on it. Yeah, you can argue that "something is being done" but the reality is nothing that means anything is being done. Myself and other programmer types have already demostrated or stated just how absurd such a short time frame is for finding and fixing anything.
Keep in mind for instance the never ending buffer overflow issues that MS takes months to "fix" without yet (to this date) resolving them. (Various months, repeatedly, over the last innumerable years, to fix it for instance in IIS - and if they cant properly fix that one bug over the span of years with months at a time being spent on it, 28 days for all of them? Heh!) If that one bug has eluded them for 7 years, while they spend months at a time attempting to fix it (unsuccessfully) then perhaps you can see that a mere 28 days to address let's say 1% of the current issues on ONE of the programs or OS's, much less everything - is an impossibility.
Oh yeah... a very respectably LARGE OS (but good?? heh!). OS/2 Warp's newest release was delivered to my doorstep last week. Barely larger than last year's release, Not much larger than the 1996 Warp 4 that everyone thinks was the last release. And it has all the features XP has and all the server components (albeit, in a non-proprietary fashion), and more networking options, and better hardware support than XP, and things that XP still doesnt have like 64 CPU per node (computer) SMP support, POSIX support, support for various connectivity options that will never make it into XP, a (still) more robust TCP/IP stack (after all, AIX's stack is FAR more robust than XP's and that's much of the basis for Warp Server's stack - including the quite real firewall that comes with Warp - while no real one comes with WinXP).
And it still runs on low end Pentiums with minimal memory (our main server is a P133 with 48MB of RAM, running SMTPD. POP3D, DNS, SQL, FTPD, HTTPD, NETBIOS NAME, TIME, SHARE and DOMAIN CONTROL - just to name the major ones. It controls the entire domain and access rights for our network - including the web servers that serve, via NetBIOS file share, millions of files a day for one of our client's adult sites). XP would choke on that hardware - much less on all those services - maybe becomes it's becoming so very good that it's the only OS with as many or less features as it's competitors but needs 2-5 times the hardware?
Then add to this. Do you really consider it becoming a very good OS when it opens 40+ ports as the default? Remember when people used to complain about the 1 NetBIOS port (#139) that Win9X opened? Now XP opens over a half dozen of them, but no one has grilled them for that. The XP services manager ensures those ports are open - period - if you want to use the 'net anyway. You can disable some of them with much hassle, while others being disabled will cripple your connectivity options even though they are not needed - and all this requires a real (ie: third party) firewall solution.
Install a Warp machine, how many ports are open? None. Not till or unless you run something that uses one. So, no... XP seems to be heading in the wrong direction. It's larger (with not much new features), it's slower [my NT4 box runs on a Pentium II 266 at 48MB albeit kinda slow... XP crawls at a near stop on twice that doing the same thing (which for me is CorelDraw and some minor graphics stuff in Photoshop)]. XP opens more ports, creates more security holes, more incompatibility and still (though to a much lesser extent than previous releases) has poor driver support (by which I mean no driver or a poorly written or crippled driver... ask Creative Labs about that one... they'll have loads to say and they are not nearly the only vendor in such a situation).
But XP is "prettier" (well that seems to be the general concensus - though I dont think so - to me it looks to childlike), so maybe that is what you are talking about when you say it is becoming a very good (meaning pretty) OS?
>And the Mac can run OS/2 in its version of Virtual PC.
On a hardware card. I was talking about software emulation of a virtual PC session. Using Connectix VPC, eventually Win users should be able to run OS/2 software - when that support is complete. And MAC users can run OS/2 or Windows on the hardware based VPC card solution. OS/2 has yet to run (since v2) on the software solution - unless you are talking about OS/2 Warp for the PowerPC which was stealthly released by IBM years ago.
You truly believe this? I've worked for (and quit) companies that MS has paid to hobble their competition's ability to compete (Comp___) for one. I've listened to Balmer and the Win95 program team lead admit they knew that Win95 wasnt near what they promised, would require many expensive upgrades they told consumers they wouldnt need and that they didnt care - and neither did our management since our employers would make a fortune off of MS's knowing lies. This, btw, is videotaped and in said company's possession. As well as 4 million dollars for the Win95 rollout as "added marketing incentive" and "marketing kits" to tell them what lies to say.
Be realistic. There's competing, and there's illegal.
Thing is, for every example you have given, there are people who do as much... that is what is odd... there seems to be no one who has done as much for satellite yet... though there are plenty of card readers and programmers - which are all simply computers.
I surely dont expect some mass market device... but I would expect someone(s) have tried or should have by now (at least on a tinkerer level...).
What's really baffling is that no one has made a hack to allow a computer to do all this. I mean, quite simply, a Dish setup nowadays is simply a bunch of MPEG streams and such. Not much more difficult than a TV tuner with an MPEG decoder and a copy protection decoder (A very simple one I'd bet).
What I can tell you is this... (1) get their top of the line - it has the best features (or review all the offerings to decide what best fits... some have some features that might be suited to you, while higher end models dont have them, but come with others... (2) dont concern yourself with reviews of their products - yet. DishNetworks in a videoconferenced "Talkback" session a few months ago indicated they are writing their own software for it, because they were tired of MS's buggy crap, inability to fix or address the issues, and slow response. Their hardware to that date, ran Win software, and soon will not anymore. The new software (theirs - not MS's) was supposed to be rolled out between now and Feb 2002, which was supposed to address many major issues these type of devices (when running the MS code) experienced. Their VP himself basically stated they thought it a great idea - they're an integration and hardware company, MS is a software company - but they were wrong as MS software was becoming a nightmare, and dealing with MS to fix it an ever larger one, so they took it upon themselves to write their own, which already exceeded MS's release.
Just FYI....
What they did detail for the various models, and such seemed a great thing. They have great expectations, though they have been slow to market with upgrades, and new features - this they also attributed to MS.
Time will tell... anyone who has seen their hardware running the new software, I'd love to hear about it. I'm waiting to see what's up before I consider it.
>Just like Mozilla and Netscape cant catch up to IE
>because AOL wont use Netscapes browser on their >own software.
BIG problem with this that many do not know. You see, when AOL started, there were really no choices in browsers... except Spry's. They thus chose Spry's browser. MS, who hasnt written their own apps since Edlin, bought it and renamed it Internet Explorer. AOL's "proprietary" system is still so dependent on it because of their original design implementations and how Rainman (proprietary AOL site building and serving system to tremendously simplify what it is) is dependent on it's Spry basis. AOL has since been doing their best to make it's services HTML compliant so they can switch off if and when they see it as opportune. AOL though, is massive. It's a tremendous, time consuming endeavor. You can note such changes already taking place (*if you have a slow connection*) when you go to an AOL area that fills grey or white and then displays it's page... that's HTML as opposed to a Rainman generated page. Slowly, they expect to have all of AOL off Rainman.
They picked the wrong company in the beginning and are stuck with them till their services are ready for an alternative.
(Or did you perhaps think they enjoyed having to release AOL 3 and early releases of 4 with IE notcompatiblewithanything 3?)
SCHecklerX makes a good point about the architectural difficulties of making something run Win apps faster... but the beauty of that is stuff like Odin (and presumably Wine - at least when compiled into Odin) call the OS's native APIs to do the actual work. *EVERY* OS for the PC is faster than WinNT/2K/XP and more capable. That leads to a tremendous chance to thrash Win9#/ME/NT/2K/XP at running it's own software. After all, Linux, eComstation and Warp can still run on 486's and run complex programs halfway decently. They run them great on Pentiums... and XP requires what???? (PII, PIII, P4???). Linux, eComstation and Warp can run in low end double digit RAM levels (in megabytes), while XP is how far into the triple digits - and to do anything decent, MS recommends what? 256MB??
If you have a car/truck engine that gets 50MPG, and puts out 500hp - but is in a 2 seater... dump it into that big van instead of it's 10MPG, 100hp engine, and you will see the difference. That's the eventual beauty of Wine and OS/2's Project Odin -far better engine under the hood to run those bloated WinWare apps. Gives Linux and OS/2 a tremendous advantage once Wine and Odin are complete.
-No OS can run Windows apps better than Windows itself.
How odd... and untrue. Recently, I tried playing a DiVX on my Win98 system. It wouldnt. "Cannot play source destination" or some other crap. Oddly, that same DiVX plays under OS/2, using WarpMedia - which uses the same Win9X codec to decode the DiVX
Looks like another OS did better there.
Then, there are the DiVX's that do play on both (which is every other one I have tried so far to be fair to Win9X...). I have the Warp system set at 1024x768. The Win system set at 640x480. I play 640x### DiVX's since it requires no scaling (and thus no extra CPU power) under Win98. Win98 manages a whopping 1-2 frames a second. Warp, scaling to full screen 1024x768, manages 30-50fps. At standard size, it can handle as high as 100fps. Leaving it synchronizing the video and audio thus results in virtually perfect playback.
Yeah, maybe a low end PII isnt what a DiVX "should" be played on - but that is only due to defficiencies in Windows... since OS/2 can obviously handle it - using the Windows DiVX codec no less!.
To play the same DiVX under Windows 98, I need to put the resolution to "1/2" (which is really 1/4 if you do the math) and manage a whopping 10 fps. Whoppee!! It turns my PII into a 486 - which is what OS/2 needs to run so pathetically playing a DiVX.
OS/2 without the nifty drivers that come with the video cards, holds its own on games like Quake II and III and newer games as well and has beaten 2K in comparisons - again, using Odin. 80-100fps on the same hardware for OS/2.
As parts of Project Odin for OS/2 and eComStation are from Wine, and the rest written from the ground up for OS/2, I'd presume that *nix users would "suffer" similar speed improvements in a number of areas as well. Yeah, some things are slower, but by and large, many are far faster - and as the Wine and Odin code progress, I can truly see Linux and OS/2 running most if not all Win apps better and faster than WinAnything. (Beating WinXP is easy as it is by far the slowest release yet... IBM Labs has a nice comparison showing it for networking. Linux thrashes it. Win2K beats it - though barely).
Sorry, but from running Win2K, XP, Win98, Linux and OS/2, I can tell you, you are very wrong. I've got plenty of identical hardware to do the tests on.
This is unfortunately not true. While I agree that the completed and not release Win9X support would indeed have helped continue OS/2's inroads into the PC market, lack of networking was not an issue. Warp3 Connect (released in late 93 I believe or early 94) had better networking support than Win95 or Win98. Including TCP/IP, NetBIOS (Win Peer to Peer lanmanager requester), Lan Client, Netware and more.
To date, it's TCP/IP stack is still second to none in the PC world (which is a statement IBM still maintains on their site - for them a very bold statement they wouldnt even consider unless not true - when for an OS they keep trying to kill), with parts of it derived from the stack that was used to host the record setting Olympic events on RS6000's (ie: borrewed from AIX) including Warp's firewall and numerous other TCP/IP daemons like routed, etc - which came from AIX as well.
Also, IBM was never slow to fix serious bugs. The SIQ problem also is not what it seems... and was addresses as soon as it became an issue, with a patch being released which was included in Warp 4, with updates for it coming out in most of the fixpacks released afterwards. IBM would release fixpacks for sometimes minimal accumulation of fixes - unlike MS... 15 for Warp 4, and sicne they supported Warp 3 up until a few months ago, something around 40 for it. Most adding additional features. Using the latest Warp 3 fix, you can upgrade it's networking to the same level as Warp 4.51 (released Nov 2000) - though not Warp 4.52 level that just came out - not much difference in that aspect though.
Win95 support did become part of the problem - but only after IBM stopped marketing OS/2. It still sold millions it's first month... and IBM dropped the ball apparently as concessions to getting Win9X at prices comparable to other OEM's instead of for retail. (see DOJ case for testimony and filings). That was the start of the true death of OS/2. No more marketing, no more preloads - though oddly, driver support took off like crazy, and now OS/2 and eComStation boast excellent driver support.
>Just like OS/2, Linux had a technical advantage (in
>some ways) over Windows up until now, but with the
>introduction of Win XP, that advantage is lessening.
I am not sure what you are talking about here. OS/2 and Linux require less hardware, support many more client connections, are more secure, have very robust web servers (both run Apache, OS/2 has 5 other actively developed http server engines), both have support for virtually every server daemon there is, both handle threads far better than any Windows incarnation. Both have highly customizable GUIs that are light years ahead of and of the WinXP/NT/98/ME/95/2K GUIs. Both support as many (or in OS/2's case, whoppingly more) CPU's per machine. Both support tremendoulsy more CPUs in a cluster.
The only area that WinXP excells in is client/workstation software - which OS/2 can now (once again) boast it can run more client software than any other OS on the planet - period - using VirtualPC, allowing it to run Windows and Linux in virtual machines under OS/2 (or eComStation).
Oh yes, and hype. XP line does indeed have much more hype.
Slashdot Mirror
Robert
(Happily running this year's release of OS/2)
Regular LED's of course are designed to do this... oh, wait, they arent!
-Robert
I guess it depends on your definition of "operating system". Almost any part of the operating system can be replaced, but at what cost? A new virtual memory manager could be swapped in, but what if that entailed kernel changes that would affect other parts of the OS? Would you then consider the VMM to be fully integrated or simply another replaceable part?
Now I'm not saying that IE goes down to that low of a level, but in a way they have integrated it into the higher level of the OS -- as a part of the various controls you can build into your app. IE itself is just another application running on your machine, but it uses the core controls that are built into the OS. As far as I'm aware, removing those controls would be like saying you no longer want radio buttons or tree-view controls. Can it be done? Yes, but at a potentially large cost to the user and development community.
>Trueexcept that Windows had those controls
>since before IE existed. It also had a File Mangler
Additionally, numerous companies and the DOJ have already proven IE can be de-"integrated" from the OS,. resulting in a speed increase, contrary to the falsified MS video to prove otherwise.
I really question this sentiment. How can you prove that something can be completely removed from the OS without modifying the source code? Are these people just proving that the IE app can be removed or that the actual browser control functionality can be completely stripped out?
>APIs that are shared by the OS dont need to
>be removed. The core APP (iexplore.exe, etc)
>and APIs in non core DLLs can be removed.
>Or, as I suggested in another post, even allowing
>the choice of a default browser that the IE WinGUI interfaceopens for URL requests
>should even satisfy such a requirement.
If they're acutally completely stripping core browser functionality, have these people proven that their actions don't effect that thousands of other Windows apps out there?
Nor is there a reason why MS couldnt write a decent file manager that doesnt need IE - or also as suggested here, at least while using IE as a file manager
See, now you're getting into territory that's wholly debateable. Is it right to ask MS to rewrite parts of the OS at a low level that may take advantage of the functionality they've built into their common controls? Maybe the file manger CAN be rewritten, but just because it can doesn't mean MS should be forced into such a task. Depending on how deep the tendrils go, redoing the file manager may be a much more dauting task than we might think.
>I think, based off the court case where that
>integration was already proven to be for anti-competitive
>reasons and deemed illegal, that it would be
>a fitting penalty - besides, there are enough
>shareware file browsers out there to prove the
>job isnt that difficult - unless MS programmers
>are really unqualified in their jobs.
I dunno, I've never studied Inside Windows 2000 or other books to see how far this stuff goes.
Right or wrong, MS has managed to get the functionality integrated with other common controls. Just pop open Visual Studio and see how easy it is to get an MFC (blech) app up and running with full browser integration. And when you take a peek at IEXPLORE.EXE in the Dependency Walker, you'll see most of the same functions that you would be using in an app that brought in a browser control.
I think the issue that MS is trying to bring up is this : IE can't just be deinstalled because much of the functionality has been incorporated into the OS via other channels.
>Or IE could be de-installed, and the API's
>that are shared by other apps could be left. Unless
>the OS is truly so poorly documented and the code
>so poorly maintained, this should be a trivial task to
>separate those APIs into those 2 baskets.
- Robert
IE itself is using that functionality and so can other apps. Removing the parts that other apps use could potentially affect a large number of users and developers. The real thing someone needs to prove is that IE and all of its functionality can be taken out and still have Windows and other apps function like nothing was missing. I personally don't know if that's the case or not.
greg
Like Windows OEM release when IE 4.0 was late for inclusion in the normal installer and was installed (if the OEM decided to, or the end user at a later date) from the second CD that came in the OEM bundle.
We still have customers who are not using the IE interface because they never ended up installing IE in that fashion to "integrate" it.
Robert
Besides, your claim implies that MS has published the IE APIs for others to use, these APIs are in now in the core API set by your example making removal of the IE executable and the IE API set in it's own DLLs still quite removable.
Unfortunately, I think your post irrelevant at best as it discusses how things could be done, not how it actually happened in the Windows world.
- Robert
Additionally, numerous companies and the DOJ have already proven IE can be de-"integrated" from the OS,. resulting in a speed increase, contrary to the falsified MS video to prove otherwise.
Add to that, Windows didnt always come with IE as the shell - and when IE 4.01/4.02 came out, the IE "integration" came (for many OEM's such as ourselves) on a second CD. It was, and is an addon.
Keep in mind, the rest of Windows components are installed on a "Check this box to install" method - there is no reason why IE cannot be installed or not installed in the same fashion. Nor is there a reason why MS couldnt write a decent file manager that doesnt need IE - or also as suggested here, at least while using IE as a file manager, block it from pulling http requests - or better yet, make an option to set and select Window's default browser and pass all URL's to it via DDE...
Also, I for one, am happy having an extensible object oriented (not object based) GUI that doesnt require IE's overhead just to open a simple drive folder. MacOS, OS/2, eComStation and some *nix desktop variants come to mind in that category.
This dead horse keeps coming up. MS knows they can easily remove IE, it was already proven in court, MS had to falsify videotaped tests to try to prove otherwise, were caught and admitted it, while the DOJ proved it could be done, and easily.
Robert - Baltimore FoodPlaces
http://www.metroactive.com/papers/metro/08.15.96 /c hain-reaction-9633.html
Wow - to think all we had to do was grab the schematics, theories, formulae and such from a bad movie!
- Robert
http://www.evworld.com/archives/interviews/balak .h tml
(follow links from article for more info)
IBM announced years ago better battery technology - it appeared in the highest end Thinkpads, and is based off Lithium and other elements (it isnt Lithium-ion), but of course, they are too expensive for say... disposable radio batteries... So, perhaps it is the economies of it that matter... after all, Lexmark, Epson and HP often sell printers at a loss - why? Because they clean up on the consumables... not that I think it is, because I have no knowledge of such, but why wouldnt the battery industry be any different? The less batteries you sell, the less you make. As new battery innovations come out (which usually are negligble and barely if at all noticeable on the "home front" AA, AAA, C, D sized cells), the prices usually go up as well... so though battery life may be increasing slightly, so is the price - at a slightly greater than inflationary rate.
Dont know, and dont care - it's really irrelevant since the item that the article incorrectly touched on was fuel cells, which ARE available and in use for things such as cell phones and other devices already. Supposedly they are more efficient, last longer, and there are methodws of recharging them (though not yet available)... so the technology does exist, and can be used to replace batteries if only more work were put into it. (currently such fuel cells are disposable instead of refillable).
- Robert
FoodPlaces: Baltimore's largest Restaurant Guide
As for such actions pissing off MS, that'd undoubtedly be the case. Just about everything seems to if it seems a snub to them. We posted a very big, nasty post complaining about their harrassment over their tactics over pirated licenses - which drove one of our suppliers to bankruptcy via lawsuits and lost customers... the supplier was innocent. MS didnt even offer an apology (Micromatix in MD was the company). And of course, the lost customers and revenue were jsut that. Anyway, shortly after our big post and complaint, our servers were attacked with the obvious intent of hacking into them. Oddly, we logged internal MS IP addresses doing the attacks. Since then, we received on a overly regular basis, notice after notice after poster after poster after letter advising us of how MS is cracking down on piracy and if we dont report everyone and anyone we know who is pirating software we will be held liable. The last one was an offer for an "agree to forget you stole our software" if we turn ourselves in. Going on over 10 mailings in as many months.
Here's the kicker... we dont run Windoze. We run MacOS 7.5, 8 and 9.04, Linux, eComStation and OS/2. We DO own 12 NT4 licenses, and 6 Win98 and 3 Win95 licenses though. We just dont use them (reformatted and re-isntalled a different OS).
I'd shudder to think of what the retribution would be if we were a mid sized to large ISP.
-Rob www.FoodPlaces.com
On our networks, logging for almost two dozen domains, the largest source of spam via "Open Relay Mail Servers" is Hotmail. These emails are being sent via other servers, and mass mailed via hotmail servers being used to relay them. Hotmail's responses to the numerous complaints? "We'll cancel that user's account..." Often though it's not the user at fault, since you dont even need a valid Hotmail address to do this. So, even with notifying them of the real problem (open servers) and showing them headers that confirm it, they do nothing. Our incoming spam would drop by over 45% if they'd fix it. - Rob
Rob
I'm an idiot, but you arre too stupid to give a single reason why? How odd.
You'd rather run a domain controller on Linux than Warp? How strange. Which one do you run? could it be the same BIND that is available for Warp? No big deal running the same server daemon since there's both the Linux version and a native up to date port available that run on Warp. Personally, I am happy with the one that comes with Warp - especially since one of our customers runs 5 adult sites so the P90 and P133 doing DNS take a beating every day (at 0-1% CPU - guess maybe they are actually quite bored).
Well, glad you like Linux. So do I. But. (1) should I care who else uses Warp? Isnt that the reason so many Linux users ignored as a valid reason why they should go to Windows? (2) Warp has more than stability on it's side. It can run virtually any Linux program, more Win32 apps than Linux as well (since it not just uses Wine, but adds many more APIs the Wine team hasnt gotten to yet that IBM did years ago, PLUS other APIs by the boatload the Project Odin team has done in the last 3 years.) and it runs them natively, by simply clicking their icon to install or run it - no configuration and installation headaches, no Win install, copy the stuff over, edit the registry crap. Just click "Setup" then when done, click the app's icon - done., (3) it makes a far better server than *ANY* PC OS to date - including Linux, sorry to say. (4) it scales to 64 CPU's per node - what else does? Every other 64 CPU setup I have seen has been clusters of 1-8 CPU per node systems. I'd rather cluster an OS that can do 64 per node. (5) I'm not advocating low end hardware - I'm advocating using whatever hardware you have properly (ie, that includes using it FULLY). At dual CPU 300MHz, my Warp system outperforms XP on 1.1GHz. At 133MHz, my Warp Server outperforms XP on 600MHz (with 3 times the RAM). My point is, I'd rather (whether the hardware is low end, high end, or inbetween), take advantage of the fact that Warp (or even Linux for that matter) will run circles around XP. One web server solution for a client to match an identical setup running Warp by us was installed for them by Microsoft. At less traffic, MS ended up installing 6 dual CPU boxes to minimize the "Server too busy" messages. I was happy with our one server, but eventually made it two. The company? Out of business, and bought out by a research company many states away. They were known as America's Doctor, online Doctor Chat and Referrals.
So, lets see, you say I'm an idiot but have yet to say why, or even how MS is making a huge step in the right direction. Yes, XP is more stable than previous releases... but that promise is a decade old - and so what? Turns out it's seeming less secure. 40+ ports open, a dozen known holes and back doors. So, where's the huge step? Maybe it's the hardware incompatibility (which to me includes crippled drivers - which has been the answer to that problem, so while most things ahve drivers, they arent nearly as capable as the Win9X versions.)
As for my other claims? Which ones? Cite one and how I need to "jump at least into the 90's" about it! Warp runs every service Linux does. It runs every service XP does. It even can run ASP stuff - on Apache 2.x no less. It runs the latest of most every server daemon out there - including Apache, MySQL, DB/2, IRC (and even the related eggdrop scripts), Domino Notes Server and more. Oooh - it doesnt run IIS - the least capable under load web server there is. So - seems my claims are quite current considering, we are (with the exception of MS branded and sold servers) talking current releases of virtually every server daemon out there (and that's not including running Linux stuff on Warp). And doing that on a current release of Warp (only a few weeks old).
So, which claim isnt true?
Keep in mind for instance the never ending buffer overflow issues that MS takes months to "fix" without yet (to this date) resolving them. (Various months, repeatedly, over the last innumerable years, to fix it for instance in IIS - and if they cant properly fix that one bug over the span of years with months at a time being spent on it, 28 days for all of them? Heh!) If that one bug has eluded them for 7 years, while they spend months at a time attempting to fix it (unsuccessfully) then perhaps you can see that a mere 28 days to address let's say 1% of the current issues on ONE of the programs or OS's, much less everything - is an impossibility.
And it still runs on low end Pentiums with minimal memory (our main server is a P133 with 48MB of RAM, running SMTPD. POP3D, DNS, SQL, FTPD, HTTPD, NETBIOS NAME, TIME, SHARE and DOMAIN CONTROL - just to name the major ones. It controls the entire domain and access rights for our network - including the web servers that serve, via NetBIOS file share, millions of files a day for one of our client's adult sites). XP would choke on that hardware - much less on all those services - maybe becomes it's becoming so very good that it's the only OS with as many or less features as it's competitors but needs 2-5 times the hardware?
Then add to this. Do you really consider it becoming a very good OS when it opens 40+ ports as the default? Remember when people used to complain about the 1 NetBIOS port (#139) that Win9X opened? Now XP opens over a half dozen of them, but no one has grilled them for that. The XP services manager ensures those ports are open - period - if you want to use the 'net anyway. You can disable some of them with much hassle, while others being disabled will cripple your connectivity options even though they are not needed - and all this requires a real (ie: third party) firewall solution.
Install a Warp machine, how many ports are open? None. Not till or unless you run something that uses one. So, no... XP seems to be heading in the wrong direction. It's larger (with not much new features), it's slower [my NT4 box runs on a Pentium II 266 at 48MB albeit kinda slow... XP crawls at a near stop on twice that doing the same thing (which for me is CorelDraw and some minor graphics stuff in Photoshop)]. XP opens more ports, creates more security holes, more incompatibility and still (though to a much lesser extent than previous releases) has poor driver support (by which I mean no driver or a poorly written or crippled driver... ask Creative Labs about that one... they'll have loads to say and they are not nearly the only vendor in such a situation).
But XP is "prettier" (well that seems to be the general concensus - though I dont think so - to me it looks to childlike), so maybe that is what you are talking about when you say it is becoming a very good (meaning pretty) OS?
-Robert
On a hardware card. I was talking about software emulation of a virtual PC session. Using Connectix VPC, eventually Win users should be able to run OS/2 software - when that support is complete. And MAC users can run OS/2 or Windows on the hardware based VPC card solution. OS/2 has yet to run (since v2) on the software solution - unless you are talking about OS/2 Warp for the PowerPC which was stealthly released by IBM years ago.
Robert
Be realistic. There's competing, and there's illegal.
Robert
I surely dont expect some mass market device... but I would expect someone(s) have tried or should have by now (at least on a tinkerer level...).
Ah well...
How odd...
Just FYI....
What they did detail for the various models, and such seemed a great thing. They have great expectations, though they have been slow to market with upgrades, and new features - this they also attributed to MS.
Time will tell... anyone who has seen their hardware running the new software, I'd love to hear about it. I'm waiting to see what's up before I consider it.
Robert
>Just like Mozilla and Netscape cant catch up to IE
>because AOL wont use Netscapes browser on their
>own software.
BIG problem with this that many do not know. You see, when AOL started, there were really no choices in browsers... except Spry's. They thus chose Spry's browser. MS, who hasnt written their own apps since Edlin, bought it and renamed it Internet Explorer. AOL's "proprietary" system is still so dependent on it because of their original design implementations and how Rainman (proprietary AOL site building and serving system to tremendously simplify what it is) is dependent on it's Spry basis. AOL has since been doing their best to make it's services HTML compliant so they can switch off if and when they see it as opportune. AOL though, is massive. It's a tremendous, time consuming endeavor. You can note such changes already taking place (*if you have a slow connection*) when you go to an AOL area that fills grey or white and then displays it's page... that's HTML as opposed to a Rainman generated page. Slowly, they expect to have all of AOL off Rainman.
They picked the wrong company in the beginning and are stuck with them till their services are ready for an alternative.
(Or did you perhaps think they enjoyed having to release AOL 3 and early releases of 4 with IE notcompatiblewithanything 3?)
Robert
If you have a car/truck engine that gets 50MPG, and puts out 500hp - but is in a 2 seater... dump it into that big van instead of it's 10MPG, 100hp engine, and you will see the difference. That's the eventual beauty of Wine and OS/2's Project Odin -far better engine under the hood to run those bloated WinWare apps. Gives Linux and OS/2 a tremendous advantage once Wine and Odin are complete.
Robert
-Then run Windows!!
-No OS can run Windows apps better than Windows itself.
How odd... and untrue. Recently, I tried playing a DiVX on my Win98 system. It wouldnt. "Cannot play source destination" or some other crap. Oddly, that same DiVX plays under OS/2, using WarpMedia - which uses the same Win9X codec to decode the DiVX
Looks like another OS did better there.
Then, there are the DiVX's that do play on both (which is every other one I have tried so far to be fair to Win9X...). I have the Warp system set at 1024x768. The Win system set at 640x480. I play 640x### DiVX's since it requires no scaling (and thus no extra CPU power) under Win98. Win98 manages a whopping 1-2 frames a second. Warp, scaling to full screen 1024x768, manages 30-50fps. At standard size, it can handle as high as 100fps. Leaving it synchronizing the video and audio thus results in virtually perfect playback.
Yeah, maybe a low end PII isnt what a DiVX "should" be played on - but that is only due to defficiencies in Windows... since OS/2 can obviously handle it - using the Windows DiVX codec no less!.
To play the same DiVX under Windows 98, I need to put the resolution to "1/2" (which is really 1/4 if you do the math) and manage a whopping 10 fps. Whoppee!! It turns my PII into a 486 - which is what OS/2 needs to run so pathetically playing a DiVX.
OS/2 without the nifty drivers that come with the video cards, holds its own on games like Quake II and III and newer games as well and has beaten 2K in comparisons - again, using Odin. 80-100fps on the same hardware for OS/2.
As parts of Project Odin for OS/2 and eComStation are from Wine, and the rest written from the ground up for OS/2, I'd presume that *nix users would "suffer" similar speed improvements in a number of areas as well. Yeah, some things are slower, but by and large, many are far faster - and as the Wine and Odin code progress, I can truly see Linux and OS/2 running most if not all Win apps better and faster than WinAnything. (Beating WinXP is easy as it is by far the slowest release yet... IBM Labs has a nice comparison showing it for networking. Linux thrashes it. Win2K beats it - though barely).
Sorry, but from running Win2K, XP, Win98, Linux and OS/2, I can tell you, you are very wrong. I've got plenty of identical hardware to do the tests on.
Robert
This is unfortunately not true. While I agree that the completed and not release Win9X support would indeed have helped continue OS/2's inroads into the PC market, lack of networking was not an issue. Warp3 Connect (released in late 93 I believe or early 94) had better networking support than Win95 or Win98. Including TCP/IP, NetBIOS (Win Peer to Peer lanmanager requester), Lan Client, Netware and more.
To date, it's TCP/IP stack is still second to none in the PC world (which is a statement IBM still maintains on their site - for them a very bold statement they wouldnt even consider unless not true - when for an OS they keep trying to kill), with parts of it derived from the stack that was used to host the record setting Olympic events on RS6000's (ie: borrewed from AIX) including Warp's firewall and numerous other TCP/IP daemons like routed, etc - which came from AIX as well.
Also, IBM was never slow to fix serious bugs. The SIQ problem also is not what it seems... and was addresses as soon as it became an issue, with a patch being released which was included in Warp 4, with updates for it coming out in most of the fixpacks released afterwards. IBM would release fixpacks for sometimes minimal accumulation of fixes - unlike MS... 15 for Warp 4, and sicne they supported Warp 3 up until a few months ago, something around 40 for it. Most adding additional features. Using the latest Warp 3 fix, you can upgrade it's networking to the same level as Warp 4.51 (released Nov 2000) - though not Warp 4.52 level that just came out - not much difference in that aspect though.
Win95 support did become part of the problem - but only after IBM stopped marketing OS/2. It still sold millions it's first month... and IBM dropped the ball apparently as concessions to getting Win9X at prices comparable to other OEM's instead of for retail. (see DOJ case for testimony and filings). That was the start of the true death of OS/2. No more marketing, no more preloads - though oddly, driver support took off like crazy, and now OS/2 and eComStation boast excellent driver support.
Rob
>Just like OS/2, Linux had a technical advantage (in
>some ways) over Windows up until now, but with the
>introduction of Win XP, that advantage is lessening.
I am not sure what you are talking about here. OS/2 and Linux require less hardware, support many more client connections, are more secure, have very robust web servers (both run Apache, OS/2 has 5 other actively developed http server engines), both have support for virtually every server daemon there is, both handle threads far better than any Windows incarnation. Both have highly customizable GUIs that are light years ahead of and of the WinXP/NT/98/ME/95/2K GUIs. Both support as many (or in OS/2's case, whoppingly more) CPU's per machine. Both support tremendoulsy more CPUs in a cluster.
The only area that WinXP excells in is client/workstation software - which OS/2 can now (once again) boast it can run more client software than any other OS on the planet - period - using VirtualPC, allowing it to run Windows and Linux in virtual machines under OS/2 (or eComStation).
Oh yes, and hype. XP line does indeed have much more hype.
- Robert