Sharepoint is free. wss 2.0 shipped with windows 2003 server, and wss3.0 is free to download.
The _only_ prerequisite is windows server. Thats it.
There is a for-pay version of sharepoint, called sharepoint portal server (new name with the just released version), but to be honest, the only people buying it are reasonable sized orgs who want to gather, search, and manage all their many different unit sharepoint webservers into one managed unit.
So no, wss3.0 (the item being talked about) does not depend on some other non-free thing called 'sharepoint'.
Before adding to the complexity of HTML they need to look at what is really wrong with HTML first and make an effort to fix those things. Fancier form elements just wouldn't be that high on my list of priorities. From my point of view, 'fancy' form elements is the biggest lack of html. Think about what you (or at least I) use js for, its 99% form validation. Why should we have to use YACL (yet another crappy language) to do something as simple as form validation and layout. This stuff should be declarative.
XSLT sucks. I used it in all my projects when it first became big and the truth is that the complexity it added was far more than it was worth. I'm sure it has value to someone but for website design it's like using a jet engine to power a lawn mower. Maybe I misunderstood, but your OP said that you wanted tags that had no inherent meaning, that you could apply transforms to with some external language. At least thats how I read what you are looking for. And thats exactly what XML/XSLT is. Yes, its crappy to work in, but no crappier than HTML and CSS, and much more powerful.
We're still having trouble getting browsers to render HTML4/CSS2 correctly and we're gonna start throwing totally new stuff out there? I agree in principle, but what can you do? It's not like anything you or I do can make IE be more compliant. But we can improve the underlying fundamentals, and just wait for the crowds to catch up.
I always attach Javascript through the head portion of my HTML and it works just fine. Again, maybe I misunderstood, but that would be disallowed under what you had proposed. I read that as only attached to external files, with observer patterns applied a'la prototype.
If you haven't ran into issues with audio or video I'd guess you haven't tried creating many multimedia web apps. Just attaching a mp3 as a link is not a solution to a case such as an XHTML/CSS/Javascript based arcade game where you need to be able to start and stop multiple sounds as events happen. I would think that maybe you're not using the right tool for the job here. No offense intended, but to my eye this is the rough equivalent of trying to build an OS out of XSLT. It's just not the right fit.
Arcade games should probably be done in a more appropriately featured platform, like Flash/Silverlight or Java/.NET. You'll be able to build them in a tiny fraction of the time, and you'll be able to work in a real development environment, and not have to deal with the random 'doesnt work' of css and javascript.
Overall, I think the HTML5 stuff is just very pragmatic. It focuses on the stuff that most of us need, and helps remove the horrid patchwork that is html/javascript/css. For web-app developers and end-users, the whole XHTML vs. HTML business is utterly pedantic, and has zero effect to providers or end-users. About the only group it helps (theoretically) is browser-makers and web-server makers, and them only once they can stop dishing/parsing old-style html.
Dont get me wrong, I find it amazing what people have been able to stretch and twist javascript/css to do, but we shouldnt have to be doing that. How about a proper forms and event based system, with a decent (ie, anything but javascript) language to do the client-side stuff. I can just picture the inventors of javascript, just laughing and laughing, thinking about all the pain they brought to future web developers, by not quite developing a real language.
And again, I'm not trying to be difficult here, I realize that we have obviously different needs and live in different worlds. All I use html for is as a GUI for business applications. I think building aracade games or heavily-multimedia stuff in html is a little crazy, given the easier/more-powerful presence of flash/silverlight.
I want to write an interactive application in JavaScript, where an audio clip of shaking leaves plays whenever the player's character walks up to a tree and grabs it. How can this be done by merely linking to an mp3 file? Fair enough, thats an application I had never thought of. I would think that would be possible with current technology, but it probably isnt elegant.
Between crappy/non-existent drivers and all the programs that are either totally incompatible with Vista, or just won't run on the 64-bit version (no Cisco VPN client? WTF??), I'm left thinking I should go back to XP. Either that, or a long waiting game for Service Pack 1. What makes you think that Vista SP1 will fix lacking drivers, poorly written 3rd party userspace apps, or Cisco VPN client? Your problem here isnt with MS and Vista, its with your software vendors.
And who sold you the hardware? What do they say when you call their support and tell them the drivers they ship with the machine doesnt work? Did you get a refund?
I now have a computer that has 1 GB more RAM than it did when running XP (grandparent is wrong on that count, XP 32-bit can't see all 4 GB of RAM because of PCI devices, etc.), but no Bluetooth, suspend-to-RAM that is completely broken,... As I asked above, what does your hardware vendor say when you call them and say that their stuff doesnt work because they didnt include the right drivers? I'd send the machine back if they cant solve it.
Go into any engineering shops, and unless they're poorly organized, they've been using x64 windows xp for years for their design & engineering work. It works just fine.
And wrt Vista-64, I think you'll find that in a competent IT shop, you will have no problems with vista once you get past the learning phase, and know what you're dealing with. Then its all pretty straightforward.
The only problem you're going to have with either XP or Vista-64 is drivers. So just buy your equipment from manufacturers who certify their equipment on x64 windows. Consider Dell, HP, and IBM all do this, you shouldnt have a problem finding equipment that works.
try to find even basic drivers for XP-64 For what its worth, this isnt a problem if you're buying new equipment. The big vendors all will support XPx64 or Vista-64 on their workstation class equipment. I know folks that have been running XP x64 for years, with very little in the way of challenges. The key seems to just buy machines that are certified for x64 windows, and there are plenty of these in the tier1 manufacturers.
Red Hat and SUSE Linux released 64-bit editions way back in 2005. No broken backwards-compatibility, no drama, no wasted billions of dollars. Are you kidding me? There is massive backwards compatibility problems on each minor version upgrade in Linux. The difference is that a large part of the software ecosystem is also open source, so it can be as simple as a recompile.
Commercial OS developers have a different reality to live in. Breaking compatibility with past commercial software that runs on your OS, even the compatibility is based on bad coding by the software vendor, is very expensive.
There's another phenomenon that plays into this. Because of things like VB and VBA in Office, the barrier to entry for software developers is _much_ lower in windows than on linux. The end result is that you have a lot of really really bad software developers, that are making a good living selling software that does an adequate job for the end-user, but is a horrid mess under the hood. This plays into the back-compat problems as well, because if one of those companies becomes very successful, then in windows fixing a bug, if it broke that software (that poorly relied on that bug), you'd alienate your users. The users (ie, those that buy windows) dont understand that their favorite craptaculous shiny email letterhead virus-maker is the problem, they just see that after a patch in windows, it doesnt work anymore.
You'd be amazed how often that happens. Its one of the unexpected side-effects of success.
There's a whole discussion you can read about this if you want, on Raymond Chen's blog. It's not quite as simple as you make it out to be.
By 2006 all Linux distributions had happy shiny 64-bit editions as a matter of routine, thanks to AMD + the sensible Linux architecture. Thats true, and it is nice. Windows has also had a happy shiny 64-bit edition since ~2003 (iirc), on XP and 2003. As long as you've got supported hardware with drivers for x64, you've been okay.
The problem with the 32-bit to 64-bit transition was never one of Windows, it was one of supported software.
One might suppose that with tens of billions of dollars and masses of talent available that Microsoft should be able to produce the most efficient, most flexible, most secure operating system in the world. If that's what the market rewarded, then it probably would. But thats not what the market rewards. In any case, if you have any involvement in IT, you should know that more money/people does not always equal a better product.
Actually, Windows 2000 is widely considered the most stable, secure Microsoft OS ever. By whom? If you work in this field, the only reason you would stay on 2000 is for legacy software or hardware that would only function there.
The only reason it _seemed_ the most secure and stable OS is the reasons I listed above. Mostly, its because the environment at the time was not nearly as hostile. The stability is largely a result of a smaller set of interactions. 2000 was pretty much corporate or enthusiast only, so you're dealing with professional management, limited hardware, limited software, limited peripherals, and a much more rigorously defined environment.
When used in an Active Directory environment, it's easy to manage. It was a dramatic improvement over an NT4 box, yes, but WinXP has a huge number more group policy options, wmi hooks, etc etc. Not to mention the built-in firewall, which although a POS compared to commercial ones, was a huge step above.
Heck, just the work done in XP SP2 on the RPC subsystem clean up, securing, and lock-down is worth the upgrade.
And XP, which you seem to like, is built on the 2k code base. Which means exactly what? Thats an incredibly meaningless statement in this context. The 2k code base was built off the NT4 code base, does that mean the NT4 code base was better? They are all built off the NT3.51 code base, does that make that one the best?
All things being equal, further maturation along a code-base is a good thing, not a bad thing.
Windows 2000 would have been the perfect opportunity for Microsoft to lock into a fixed release cycle. It was good enough... By what measure was 2000 'good enough'?
Think about all the fundamental architectural changes that have gone into windows since then, across XP, 2003/xpsp2, vista. And then there's the shift from 32-bit to 64-bit, with all the breaking of back-compat that they do with that transition.
2000 was horribly broken in so many ways, tremendously insecure, and just barely manageable in a corporate environment.
The only reason it 'seems good enough' is that it was the first windows box from the NT codebase that had a pretty interface. So it looked as good as windows 95, but had the stability of the nt codebase. It also was pretty much corporate-only, so you didnt have to deal with the glut of consumer-peripheral trash and their outsourced driver writers, not to mention the legacy software written for 9x systems.
It just seemed decent because it was such an improvement over its predecessor for most people (most people went win9x->win2000), and the overall security environment wasnt what it was today.
Unfortunately for everyone, they got lazy and sat on their monopoly, and fell behind OS X (and more recently Linux) in terms of security, visual quality, and ease of use. Now they're at a point where they've got a bug-ridden half-featured OS released to try and catch up. They need to get back to the stability of Windows 2000 before they can lock into a stable release cycle. Yeah, because trying to massively improve the system, being so ambitious in their work that they overshot and had to 'reset', being willing to tackle the back-compat problems with the 32-bit to 64-bit transition, and completely re-engineering all of their internal processes to incorporate security-aware programming and processes.
Phew! Those lazy guys... those MS developers must have just spend the last 5 years lazing around on couches drinking capuccinos.
Vista is the transitional OS, where they are breaking most of the back-compat bonds with the past.
Vista is a massive re-engineering under the hood, new schedulers, new IO, new TCP/IP, new security modeling, new memory management. The changes are pretty damn massive, but most of the changes arent what you see. They just make the system more robust over the long hault.
The big change though is the transition from 32-bit to 64-bit. And in that transition, MS is breaking a huge amount of back-compat.
I think what you'll find, in 3 years, is that Vista ended up being one of those silent improvements. Yeah, nothing fundamentally changed for the non-technical end-user, but the guts were very fundamentally re-done. And so the system is much more robust, more stable, less prone to security issues, and in general, 'just-works' much better.
But the key thing is the 32-bit to 64-bit transition, and the (finally) breaking of 10 years of back-compat problems.
If MS could figure out how to run both 32 and 64 bit code at the same time [such as how at least one other major OS can], it would greatly aid both developers and end-users. Have you never run an x64 version of Windows? Thats how it works now. In fact, you'll notice that when you install software, if its 64-bit, it goes in Program Files, and if its 32-bit, it goes into Program Files (x86).
Most 32-bit code runs just fine on 64-bit windows. Where it doesnt falls nicely into a few categories:
1. Systems software.
2. Poorly Designed Software.
3. Drivers.
1 and 3 are no-brainers. 2 is the cause of much of what you see. For example, how many anti-virus systems are there where the actual anti-virus code works just fine on 32-bit and 64-bit systems, but their management software (just GUI and messaging) wont run on 64-bit systems. This is a perfect example of poorly built software.
Unsigned code and antitrust lawsuits do have connection if you have to pay to get singed code and only m$ can say what gets signed. M$ is doing this right now some types of drivers for vista. You're badly misrepresenting the situation.
You dont pay MS to get the code-signing cert, you pay a certification company, like Verisign, etc. And MS has no involvement whatsoever, in any way, with what code signing certs get issued.
MS is NOT restricting who can get code-signing certs for drivers. It doesnt work that way, at all. Read up on the process.
The only involvement MS has, as the root signing cert, is that they can revoke child certs, and ship that revocation list into windows with the regular updates. As far as I know, this has only been done a few times ever, and in every case due to fraudulently obtained certs, with this used to potentially distribute malware.
I think we're looking at completely different sets of problems. The stuff you're talking about is designer-targeted issues, which I dont care so much about. At least some of the things in WHATWG's HTML5 is focused on web-app developers, which is far more important in my world.
Not that one world is more important, but they're completely different sets of problems, and probably will have completely different solutions.
XHTML is still bettr than HTML in that it's a lot easier to parse. IMHO that is a good thing as it makes it easier for browsers to be smaller and bug free. The point that any device can also render HTML is moot because it still makes the devices work harder and the majority of websites still don't use XHTML because the majority of web designers are trapped in 1997. The majority of websites still just can't easily be tweaked to work on multiple types of devices to any decent level of usability - many don't even work on different browsers or screen sizes properly.
Given what you seem to be looking for, why dont you just go with shipping pure XML and XSLT, and bypass the whole situation? Ship a different style sheet for each client type. It sounds like what you're really looking for is the maxima of complete separation of content and presentation. Doesnt XML/XSLT give you that?
We shouldn't be adding a bunch of specialized tags - we need to make it easier to define roles for tags, using CSS and Javascript, that can change as needed. Why not just make tags as optional/flexible as setting the class of a tag. So I can use MyText instead of MyText when I want to. It'd just lead to code being easier to read. Tag attributes such as href, src, rel, etc would be easier if CSS was expanded so that they could be set in that way.
Again, it sounds like you just want XML and XSLT, where tags/elements dont mean anything in and of itself, and you use some other scripting/presentation/transformation process to make it present well.
It looks like they still let Javascript ans styles be written inline so it can't be that good a proposal. It just leads to sloppy code and issues like XSS attacks.
This is one of those ideas that sounds great up front, but once you think about it for a little while, has fundamental side-effects that could be very problematic.
For example, many web apps, or web-sites that produce pages dynamically based on some condition, may really struggle with no inline js. I'll give you that it would really help mitigate XSS attacks, but I'm not sure the protection is worth the cost.
I guess it might be possible to produce your dynamic js in a separate output stream, cache it and then have it respond from the page request... but man does that seem over-complicated. I'm going to continue to think about it though, as getting all inline JS off pages would have some nice side-effects, but I'm just worried that the negative side-effects might be worse.
What we really need is a correct way to specify code that responds to specific browsers and browser versions, to specific disabilities, and to specific window sizes.
You have that now, just produce different HTML/XHTML based on the client, in your server-side code.
More widgets being available is great but lets fix some of the fundamental issues first.
For many of us, the current state of form tags/elements/widgets IS a fundamental issue, and much more important than audio/video or layout/design problems. In my mind, if you want audio/video, make a link to.mp3 or.mpeg files. Easy as pie, and doesnt require anything new. And an infinitely better user-experience than playing through some horrible flash client, which wont let you re-size or save.
The primary purpose many of us use HTML for is as a GUI for our applications. Again, I'm not saying thats the only reason, but I think there's plenty of room for improvement in that space, without limit
As an IT manager, I have a nice Sony Vaio Well, thats your first mistake.:)
Seriously though:
Either you have shared video RAM, which is definitely a generation or more behind in performance compared to a dedicated RAM high speed card, or you have x amount of of Video RAM, which cannot be upgraded. Plenty of laptops have 256 or 512MB onboard (not shared) memory, with current-gen engines.
You're right you cant upgrade it.
hese are almost always less RAM then you can get on the newest cards, and even if you pop for the ultimate primo top of the line king of the hill video setup for your laptop, within six months, you existing on board Video will be blown away by the latest greatest available, with no opportunity to upgrade. Do you really buy the latest, greatest video card every 6-months? That is a lot of money, and you're way out on the bleeding edge with that kind of a computing lifestyle.
And even if you dont have the absolutely greatest video card in 6-months, who cares? Is your goal to play video games or show up your neighbors? Because you can certainly get a good enough card to play the latest and greatest video games.
Add to that the fact that you only get ONE adapter on a laptop Not true at all. The laptop chassis itself may only have on output, but its a dual-head card, and as soon as you plug it into the docking-station/port-replicator, you get 2 outputs. And if you have a docking station with the right video card, you can have a 4-monitor system at your desk (though the docking station card will be pci, so no big 3d).
Laptops use the same video connector as desktops, so whatever 24" screen you like on your desktop, then use it on your laptop.
Having direct-attached very-fast, very-large disk storage is out of the reach of laptops, but you can get pretty close. External drivers (raid arrays as well) that connect via firewire, usb2, or gigabit ethernet.
And plenty of laptops have available as an option high-end 3d cards with 512MB of onboard memory. You cant get quite as high-end as the highest-end desktop cards, and you cant do SLI, but you can have a pretty damn good gaming video card if you want in a laptop.
How much storage space do you have in your laptop? I haven't seen a single laptop that provides more than 250GB HDD space (not that I've been particularly looking). That's not nearly enough for what many people use desktops for, and that isn't a problem I see getting solved any time soon. I dont know, I've never seen a use case where a person needs more than 250GB of storage space that is also mobile. I can understand if they want to store a ton of pr0n/movies/music that they get off bittorrent, but do all those movies need to travel with them while they go?
If not, then just use a NAS at home, or a big fat external drive at home.
I'm sure there are some people with that situation, but it doesnt seem very common. For those edge-cases, they're just not a good person for a laptop.
I personally like my desktop because if I spill water on the keyboard it is a $15.00 repair, on a laptop that could easily run to a full replacement. This is a non-issue with modern laptops.
Pour all the water or liquid you want into the keyboard, there's no path for the water to take into the inside, it just runs out the side, or bubbles over the top. They're specifically designed to handle this nowadays.
Thats why you have docking stations and port-replicators.
You can have exactly the same set of peripherals, dual-monitors, big keyboards, great sound, etc with your laptop. It just all plugs into the docking station.
This way you get portability, and you also get ergonomic when you're at your desk.
go through all the constant hassle of docking and undocking Because applying pressure onto the top of the laptop (to dock), or pressing the 'Undock' button (to un-dock) is hard?
It is more and more common now for people, common users, to want to take their own home shot video, and edit it down, enhance it with audio, etc, and render it for DVD burning. With the greater availability of HD cameras for home use, the need for large amounts of ram and fast multi-core processors will increase. Many of the things people are wanting to use computers at home for require more than can be loaded on a laptop, even with adding peripheral harddrives for added storage. None of these things require an 8-way machine with 16+ GB of ram and a RAID subsystem.
We get laptops all the time with dual-core processors, 4GB of memory, and 256MB Quadro (or better) video cards. These are mainstream dell latitude & precision boxes.
One of these would do everything you talk about just fine.
And heck, its only been a few months since you could put dual-quad-core processors in a machine for less than $10,000.
Your desktop at home has your regular 2 big LCDs, you've got a regular keyboard & mouse, and you've got ethernet, and your great speakers, and an external hdd all hooked up to the docking-station/port-replicator.
So when you're mobile, you're working on the laptop, and its a good trade-off between mobility and usability.
When you're home, it takes 1 second to dock, and then you've got exactly the same environment you would have with your desktop.
I love web standards. They're made from this naive academic ivory tower with absolutely no recognition of the commercial or aesthetic requirements of the actual web as it's used by... everybody else, I guess. I cant speak to whether some of the earlier w3c specs were like that, but I dont think thats an accurate representation of this group.
WHATWG which this is based on is basically the Safari folks, the Opera folks, and the Mozilla folks. Most of those are pretty pragmatic. The Opera folks in particular I'm excited to see involved. They do some really neat work.
And the w3c group is alot of the same people as the WHATWG group.
Plus the group is structured to be more open than typical, allowing many outside experts to participate.
Did you read the proposal, or anything around WHAT-WG's HTML5?
It's actually incredibly sensible, and is a very practical and natural extension of what we're doing with HTML now.
It has very little to do with browser bugs, or even web sites per-se. It's more about adding features to more naturally support web 'apps'.
Read up on it, it actually makes a lot of sense.
I just hope it can make some progress, but given that it was started by Mozilla, Apple and Opera, the people making the best browsers out there, it may actually have a chance of being supported.
Slashdot Mirror
Are you kidding me?
Sharepoint is free. wss 2.0 shipped with windows 2003 server, and wss3.0 is free to download.
The _only_ prerequisite is windows server. Thats it.
There is a for-pay version of sharepoint, called sharepoint portal server (new name with the just released version), but to be honest, the only people buying it are reasonable sized orgs who want to gather, search, and manage all their many different unit sharepoint webservers into one managed unit.
So no, wss3.0 (the item being talked about) does not depend on some other non-free thing called 'sharepoint'.
Arcade games should probably be done in a more appropriately featured platform, like Flash/Silverlight or Java/.NET. You'll be able to build them in a tiny fraction of the time, and you'll be able to work in a real development environment, and not have to deal with the random 'doesnt work' of css and javascript.
Overall, I think the HTML5 stuff is just very pragmatic. It focuses on the stuff that most of us need, and helps remove the horrid patchwork that is html/javascript/css. For web-app developers and end-users, the whole XHTML vs. HTML business is utterly pedantic, and has zero effect to providers or end-users. About the only group it helps (theoretically) is browser-makers and web-server makers, and them only once they can stop dishing/parsing old-style html.
Dont get me wrong, I find it amazing what people have been able to stretch and twist javascript/css to do, but we shouldnt have to be doing that. How about a proper forms and event based system, with a decent (ie, anything but javascript) language to do the client-side stuff. I can just picture the inventors of javascript, just laughing and laughing, thinking about all the pain they brought to future web developers, by not quite developing a real language.
And again, I'm not trying to be difficult here, I realize that we have obviously different needs and live in different worlds. All I use html for is as a GUI for business applications. I think building aracade games or heavily-multimedia stuff in html is a little crazy, given the easier/more-powerful presence of flash/silverlight.
And who sold you the hardware? What do they say when you call their support and tell them the drivers they ship with the machine doesnt work? Did you get a refund? I now have a computer that has 1 GB more RAM than it did when running XP (grandparent is wrong on that count, XP 32-bit can't see all 4 GB of RAM because of PCI devices, etc.), but no Bluetooth, suspend-to-RAM that is completely broken,
He's really not full of shit.
Go into any engineering shops, and unless they're poorly organized, they've been using x64 windows xp for years for their design & engineering work. It works just fine.
And wrt Vista-64, I think you'll find that in a competent IT shop, you will have no problems with vista once you get past the learning phase, and know what you're dealing with. Then its all pretty straightforward.
The only problem you're going to have with either XP or Vista-64 is drivers. So just buy your equipment from manufacturers who certify their equipment on x64 windows. Consider Dell, HP, and IBM all do this, you shouldnt have a problem finding equipment that works.
Commercial OS developers have a different reality to live in. Breaking compatibility with past commercial software that runs on your OS, even the compatibility is based on bad coding by the software vendor, is very expensive.
There's another phenomenon that plays into this. Because of things like VB and VBA in Office, the barrier to entry for software developers is _much_ lower in windows than on linux. The end result is that you have a lot of really really bad software developers, that are making a good living selling software that does an adequate job for the end-user, but is a horrid mess under the hood. This plays into the back-compat problems as well, because if one of those companies becomes very successful, then in windows fixing a bug, if it broke that software (that poorly relied on that bug), you'd alienate your users. The users (ie, those that buy windows) dont understand that their favorite craptaculous shiny email letterhead virus-maker is the problem, they just see that after a patch in windows, it doesnt work anymore.
You'd be amazed how often that happens. Its one of the unexpected side-effects of success.
There's a whole discussion you can read about this if you want, on Raymond Chen's blog. It's not quite as simple as you make it out to be. By 2006 all Linux distributions had happy shiny 64-bit editions as a matter of routine, thanks to AMD + the sensible Linux architecture. Thats true, and it is nice. Windows has also had a happy shiny 64-bit edition since ~2003 (iirc), on XP and 2003. As long as you've got supported hardware with drivers for x64, you've been okay.
The problem with the 32-bit to 64-bit transition was never one of Windows, it was one of supported software. One might suppose that with tens of billions of dollars and masses of talent available that Microsoft should be able to produce the most efficient, most flexible, most secure operating system in the world. If that's what the market rewarded, then it probably would. But thats not what the market rewards. In any case, if you have any involvement in IT, you should know that more money/people does not always equal a better product.
The only reason it _seemed_ the most secure and stable OS is the reasons I listed above. Mostly, its because the environment at the time was not nearly as hostile. The stability is largely a result of a smaller set of interactions. 2000 was pretty much corporate or enthusiast only, so you're dealing with professional management, limited hardware, limited software, limited peripherals, and a much more rigorously defined environment. When used in an Active Directory environment, it's easy to manage. It was a dramatic improvement over an NT4 box, yes, but WinXP has a huge number more group policy options, wmi hooks, etc etc. Not to mention the built-in firewall, which although a POS compared to commercial ones, was a huge step above.
Heck, just the work done in XP SP2 on the RPC subsystem clean up, securing, and lock-down is worth the upgrade. And XP, which you seem to like, is built on the 2k code base. Which means exactly what? Thats an incredibly meaningless statement in this context. The 2k code base was built off the NT4 code base, does that mean the NT4 code base was better? They are all built off the NT3.51 code base, does that make that one the best?
All things being equal, further maturation along a code-base is a good thing, not a bad thing.
Think about all the fundamental architectural changes that have gone into windows since then, across XP, 2003/xpsp2, vista. And then there's the shift from 32-bit to 64-bit, with all the breaking of back-compat that they do with that transition.
2000 was horribly broken in so many ways, tremendously insecure, and just barely manageable in a corporate environment.
The only reason it 'seems good enough' is that it was the first windows box from the NT codebase that had a pretty interface. So it looked as good as windows 95, but had the stability of the nt codebase. It also was pretty much corporate-only, so you didnt have to deal with the glut of consumer-peripheral trash and their outsourced driver writers, not to mention the legacy software written for 9x systems.
It just seemed decent because it was such an improvement over its predecessor for most people (most people went win9x->win2000), and the overall security environment wasnt what it was today. Unfortunately for everyone, they got lazy and sat on their monopoly, and fell behind OS X (and more recently Linux) in terms of security, visual quality, and ease of use. Now they're at a point where they've got a bug-ridden half-featured OS released to try and catch up. They need to get back to the stability of Windows 2000 before they can lock into a stable release cycle. Yeah, because trying to massively improve the system, being so ambitious in their work that they overshot and had to 'reset', being willing to tackle the back-compat problems with the 32-bit to 64-bit transition, and completely re-engineering all of their internal processes to incorporate security-aware programming and processes.
Phew! Those lazy guys
Close but not quite.
Vista is the transitional OS, where they are breaking most of the back-compat bonds with the past.
Vista is a massive re-engineering under the hood, new schedulers, new IO, new TCP/IP, new security modeling, new memory management. The changes are pretty damn massive, but most of the changes arent what you see. They just make the system more robust over the long hault.
The big change though is the transition from 32-bit to 64-bit. And in that transition, MS is breaking a huge amount of back-compat.
I think what you'll find, in 3 years, is that Vista ended up being one of those silent improvements. Yeah, nothing fundamentally changed for the non-technical end-user, but the guts were very fundamentally re-done. And so the system is much more robust, more stable, less prone to security issues, and in general, 'just-works' much better.
But the key thing is the 32-bit to 64-bit transition, and the (finally) breaking of 10 years of back-compat problems.
Most 32-bit code runs just fine on 64-bit windows. Where it doesnt falls nicely into a few categories:
1. Systems software.
2. Poorly Designed Software.
3. Drivers.
1 and 3 are no-brainers. 2 is the cause of much of what you see. For example, how many anti-virus systems are there where the actual anti-virus code works just fine on 32-bit and 64-bit systems, but their management software (just GUI and messaging) wont run on 64-bit systems. This is a perfect example of poorly built software.
You dont pay MS to get the code-signing cert, you pay a certification company, like Verisign, etc. And MS has no involvement whatsoever, in any way, with what code signing certs get issued.
MS is NOT restricting who can get code-signing certs for drivers. It doesnt work that way, at all. Read up on the process.
The only involvement MS has, as the root signing cert, is that they can revoke child certs, and ship that revocation list into windows with the regular updates. As far as I know, this has only been done a few times ever, and in every case due to fraudulently obtained certs, with this used to potentially distribute malware.
Not that one world is more important, but they're completely different sets of problems, and probably will have completely different solutions.
XHTML is still bettr than HTML in that it's a lot easier to parse. IMHO that is a good thing as it makes it easier for browsers to be smaller and bug free. The point that any device can also render HTML is moot because it still makes the devices work harder and the majority of websites still don't use XHTML because the majority of web designers are trapped in 1997. The majority of websites still just can't easily be tweaked to work on multiple types of devices to any decent level of usability - many don't even work on different browsers or screen sizes properly.
Given what you seem to be looking for, why dont you just go with shipping pure XML and XSLT, and bypass the whole situation? Ship a different style sheet for each client type. It sounds like what you're really looking for is the maxima of complete separation of content and presentation. Doesnt XML/XSLT give you that?
We shouldn't be adding a bunch of specialized tags - we need to make it easier to define roles for tags, using CSS and Javascript, that can change as needed. Why not just make tags as optional/flexible as setting the class of a tag. So I can use MyText instead of MyText when I want to. It'd just lead to code being easier to read. Tag attributes such as href, src, rel, etc would be easier if CSS was expanded so that they could be set in that way.
Again, it sounds like you just want XML and XSLT, where tags/elements dont mean anything in and of itself, and you use some other scripting/presentation/transformation process to make it present well.
It looks like they still let Javascript ans styles be written inline so it can't be that good a proposal. It just leads to sloppy code and issues like XSS attacks.
This is one of those ideas that sounds great up front, but once you think about it for a little while, has fundamental side-effects that could be very problematic.
... but man does that seem over-complicated. I'm going to continue to think about it though, as getting all inline JS off pages would have some nice side-effects, but I'm just worried that the negative side-effects might be worse.
For example, many web apps, or web-sites that produce pages dynamically based on some condition, may really struggle with no inline js. I'll give you that it would really help mitigate XSS attacks, but I'm not sure the protection is worth the cost.
I guess it might be possible to produce your dynamic js in a separate output stream, cache it and then have it respond from the page request
What we really need is a correct way to specify code that responds to specific browsers and browser versions, to specific disabilities, and to specific window sizes.
You have that now, just produce different HTML/XHTML based on the client, in your server-side code.
More widgets being available is great but lets fix some of the fundamental issues first.
For many of us, the current state of form tags/elements/widgets IS a fundamental issue, and much more important than audio/video or layout/design problems. In my mind, if you want audio/video, make a link to .mp3 or .mpeg files. Easy as pie, and doesnt require anything new. And an infinitely better user-experience than playing through some horrible flash client, which wont let you re-size or save.
The primary purpose many of us use HTML for is as a GUI for our applications. Again, I'm not saying thats the only reason, but I think there's plenty of room for improvement in that space, without limit
Seriously though: Either you have shared video RAM, which is definitely a generation or more behind in performance compared to a dedicated RAM high speed card, or you have x amount of of Video RAM, which cannot be upgraded. Plenty of laptops have 256 or 512MB onboard (not shared) memory, with current-gen engines.
You're right you cant upgrade it. hese are almost always less RAM then you can get on the newest cards, and even if you pop for the ultimate primo top of the line king of the hill video setup for your laptop, within six months, you existing on board Video will be blown away by the latest greatest available, with no opportunity to upgrade. Do you really buy the latest, greatest video card every 6-months? That is a lot of money, and you're way out on the bleeding edge with that kind of a computing lifestyle.
And even if you dont have the absolutely greatest video card in 6-months, who cares? Is your goal to play video games or show up your neighbors? Because you can certainly get a good enough card to play the latest and greatest video games. Add to that the fact that you only get ONE adapter on a laptop Not true at all. The laptop chassis itself may only have on output, but its a dual-head card, and as soon as you plug it into the docking-station/port-replicator, you get 2 outputs. And if you have a docking station with the right video card, you can have a 4-monitor system at your desk (though the docking station card will be pci, so no big 3d).
Laptops use the same video connector as desktops, so whatever 24" screen you like on your desktop, then use it on your laptop.
:)
Having direct-attached very-fast, very-large disk storage is out of the reach of laptops, but you can get pretty close. External drivers (raid arrays as well) that connect via firewire, usb2, or gigabit ethernet.
And plenty of laptops have available as an option high-end 3d cards with 512MB of onboard memory. You cant get quite as high-end as the highest-end desktop cards, and you cant do SLI, but you can have a pretty damn good gaming video card if you want in a laptop.
Of course it wont be cheap.
If not, then just use a NAS at home, or a big fat external drive at home.
I'm sure there are some people with that situation, but it doesnt seem very common. For those edge-cases, they're just not a good person for a laptop.
Pour all the water or liquid you want into the keyboard, there's no path for the water to take into the inside, it just runs out the side, or bubbles over the top. They're specifically designed to handle this nowadays.
Thats why you have docking stations and port-replicators.
You can have exactly the same set of peripherals, dual-monitors, big keyboards, great sound, etc with your laptop. It just all plugs into the docking station.
This way you get portability, and you also get ergonomic when you're at your desk.
What kind of equipment have you been using?
I've been using Dell & HP laptops with docking stations for the best part of 10 years, and have never even heard of or seen anything like this.
Were you using the actual manufacturer port-replicator/docking-stations? Or were you using those lame Belking USB "docking stations"?
We've found that the docking stations (from dell & HP at least) will often outlast the laptops.
Besides, even with a USB hub, you have a minimum of 5 connections to make and un-make every time you move:
- power
- ethernet
- video
- usb hub
- audio
We get laptops all the time with dual-core processors, 4GB of memory, and 256MB Quadro (or better) video cards. These are mainstream dell latitude & precision boxes.
One of these would do everything you talk about just fine.
And heck, its only been a few months since you could put dual-quad-core processors in a machine for less than $10,000.
Most of these things aren't relevant.
Your desktop at home has your regular 2 big LCDs, you've got a regular keyboard & mouse, and you've got ethernet, and your great speakers, and an external hdd all hooked up to the docking-station/port-replicator.
So when you're mobile, you're working on the laptop, and its a good trade-off between mobility and usability.
When you're home, it takes 1 second to dock, and then you've got exactly the same environment you would have with your desktop.
WHATWG which this is based on is basically the Safari folks, the Opera folks, and the Mozilla folks. Most of those are pretty pragmatic. The Opera folks in particular I'm excited to see involved. They do some really neat work.
And the w3c group is alot of the same people as the WHATWG group.
Plus the group is structured to be more open than typical, allowing many outside experts to participate.
I'd be real surprised if MS didnt end up implementing the WHATWG's HTML5. Thats of course assuming that everything around HTML5 works out as expected.
But the functionality included in HTML5 are so damned compelling from a web-app developer's standpoint, that I expect it will create alot of demand.
Of course, MS' release cycle timing of IE can be pretty strange, and that may be one of their biggest challenges.
In any case, this HTML5 stuff is compelling. I sure do hope it makes some good progress.
Did you read the proposal, or anything around WHAT-WG's HTML5?
It's actually incredibly sensible, and is a very practical and natural extension of what we're doing with HTML now.
It has very little to do with browser bugs, or even web sites per-se. It's more about adding features to more naturally support web 'apps'.
Read up on it, it actually makes a lot of sense.
I just hope it can make some progress, but given that it was started by Mozilla, Apple and Opera, the people making the best browsers out there, it may actually have a chance of being supported.