Google vs. Microsoft On the Desktop

Michael_Curator writes "Gary Edwards, president of the now-defunct Open Document Foundation, helps sort out the challenges Google faces displacing Microsoft on the desktop, pitting the strengths of Microsoft's proprietary stack against the developer candy that HTML 5 represents."

Comments

Take away the cloud

[Darkness404]

What I think would be best for Google would be to fork a version of OOo to include "Save to the cloud" support and integration with Google Docs. Along with integration with every e-mail client by using perhaps HTML e-mail or a plugin to enable Google Docs support. Create an iPhone app, plugins for MS Office, make it easy for anyone with any program to access and use Google Docs and it will succeed.

Re:Take away the cloud

[aaron.axvig]

HTML is just another layer of abstraction. It could just as well be Java, or .NET CLR, or cross-compiled C++ (GIMP). There is nothing amazing about applications in a browser, it is not necessary, and while it is convenient at times (at a computer that is not your own), when available a native code app will usually do the same job but "better".

As far as syncing, there is nothing stopping native apps from syncing to "the cloud". In fact, there is the Outlook Connector for Windows Live Mail and the Office Live tool for Word XP, 2003, and 2007. Also see IMAP and POP3. Oh, basically anything that doesn't go over port 80.

Browser is not a necessity for productivity. Handy in cases, yes. Disclosure: I'm currently interning at MS.

Re:Take away the cloud

[orngjce223]

I *already* compute in the cloud. Except for the stuff that needs high security (like stuff I'd like to actually claim as intellectual property someday) - uploading that to a website that in its EULA claims the right to read your data is just stupid.

Then again, the reason I *do* do so is because I use five different computers and if it isn't the cloud, it's the sneakernet and I'm notorious for losing USB-flashdrives.

Re:Take away the cloud

[docbrody]

But Microsoft people have a good point about the cloud. Forget speed, think about reliability. And by reliability of the cloud, I actually mean reliability of your internet connection.

I think it will be a long time before the internet/cloud can compete with local internal storage. So for Google to compete, cloud features are an awesome additional feature, but to really succeed, I think they need to be able to go toe-to-toe with Microsoft on the desktop without requiring an internet connection.

Re:Take away the cloud

[fractoid]

What I think would be best for Google would be to fork a version of OOo to include "Save to the cloud" support and integration with Google Docs.

"Save to the cloud"? Oh god, make the buzzing stop! You mean "add an option to OpenOffice to save your files to a remote server". Calling it "the cloud" is like calling the contents of your hard drive "cyberspace".

Re:Take away the cloud

[drinkypoo]

It really is 'the cloud' though; face it, if you save to google you're saving to a cluster. You have no idea where your data is and you don't care. To say you're saving it to a server is a bit disingenuous. You might as well just draw the good old cloud and lightning bolt ala the network diagrams of old and leave it at that, in most situations.

Re:Take away the cloud

[miffo.swe]

Getting rid of stupid clients would be godsent for any admin in the world. Having all applications in the browser would be a huge step forward.

You try, getting three different clients working against a database from the same vendor working properly. They all crave different versions of dotnet, java or whatnot and any new version of the client software demands countless hours of testing just about every possible combination of apps. Upgrades are pure nightmare. Couple this with locked down desktops, profiles that has to be managed and policies that needs hard testing before you alter a single setting.

Getting rid of all those problems alone would be worth serious money for any company. Added benefit would be that backend services would be totally decoupled from what OS the client runs. Microsoft will fight this for all they are worth.

Re:Take away the cloud

[lanswitch]

most people don't know what a server or cloud is. they don't even care. all they want is that they can find their documents.
i guess it will go like this: "if i store it on the cloud/server/whatever, it's not always there, but if i put it in 'My Documents', it's always available. so i'll store it on 'My Documents""

Re:Take away the cloud

[MaggieL]

Maybe. Who will claim the win when Android tops iPhone in 2012? RMS?

Re:Take away the cloud

[Tablizer]

Getting rid of stupid clients would be godsent for any admin in the world. Having all applications in the browser would be a huge step forward.

Deployability seems to continue to trump a richer GUI in organizations. To-desktop installation comes with tons of headaches. Many organizations have made it require administrative access to install new apps so that people don't download executable trash from the web. However, this also makes it difficult to download legitimate apps. Finer-granularity permissions management is not yet perfected.

     

Re:Take away the cloud

[MaggieL]

You're interning at MSFT? Or internalizing? Both?

Re:Take away the cloud

[drinkypoo]

And then their hard drive takes a gigantic shit and they learn the value of a multi-tiered approach.

On the other hand, always-on internet access is real for some people already.

Re:Take away the cloud

[syphax]

Or you use DropBox which seamlessly syncs your local data with the "cloud" and whatever other machines you have.

It's pretty sweet. Every photo of my kids I upload from a camera to my PC (OK, Mac) appears on the grandparents' PC (OK, that's a Mac too) shortly thereafter (Yes, I know I could just cron a unison job across a SSH tunnel to get a similar effect, but a) I'm too old for that crap and b) I want auto offsite backup).

Re:Take away the cloud

[shawb]

Alternatively, they could use the "reasoning" of "You mean, if I downloaded up to the intarnetz I can get my stuff from school, my friend's house, or my iphone without having to figure out all those scary cables and thumbdrivamagiggers?"

If your average person can't get access to their files because their internet connection is down... they just go do something else for a while. Your typical (or at least stereotypical) slashdotter should be able to find an alternative way to get internet access if the files are that important... neighbor's unsecured wifi, tethered cell-phone, bringing your laptop to someplace that does have access, or even more, um, exotic alternatives.

Re:Take away the cloud

[chabotc]

"Save to remote server" ?

Personally I think that a end-user would be slightly confused by such mumbo jumbo, I mean, do you really know a lot of non-technies that know what a 'cervaaar' is?

If such an option were to be added, please let it be called "Save it to Google docs" :)

Re:Take away the cloud

[Anonymous Coward]

All platforms are "just another layer of abstraction". Windows for example. Of cource everything runs better if you write it all out in machine code and key it into memory directly, but that isn't practical in any sense. The purpose of a platform is to be the most efficient, direct translation of high level to low level code. .

The reason that web applications have an advantage is primarily the platform on which they sit. They're compatible with all operating systems, portable, simple to build, and really really light weight. As performance ramps up (as it most certainly is with google chrome) and more things are possible in a browser, we will see greater and greater market dominance of web applications. This is why web standards are so important.

One could argue that you're imposing an artifical lifespan on your application by writing it in something like C, as you are directly tieing it to the one and only software environment in which it runs.

Re:Take away the cloud

[chabotc]

The classic model where people only worked on and with local docs and programs is long gone for the newer generations, and without internet their 'computer is broken', since their facebook, favorite flash game, IM, email and home/search page all give weird error messages.

That, plus the benefits of always having your documents with you no matter what computer and operating system they are using or what location they are at, and the ability to collaborate, share and publish are pretty strong arguments against the local 'My Documents' type model.

Re:Take away the cloud

[fractoid]

If such an option were to be added, please let it be called "Save it to Google docs" :)

Yes please. We have enough mumbo jumbo, as you put it, without inventing new wannabe-cool terms for things.

Re:Take away the cloud

[hairyfeet]

Actually even my 67 year old clueless dad has a USB external with one touch backup. Those things are really dirt cheap now and most folks have been burnt at least once in the past and are naturally a little spooked about losing all their stuff.

Which kinda spooks me about this whole "push to the cloud" thing. Not only do you have the whole privacy issue, because you have no idea who is looking at your data, but lets be honest here, even Google has occasionally just went "oops, sorry about that" when it has come to folks data. I know I have Gmail go offline for a day or two at a time, and what if one of those days I had been on the road and having that data was "mission critical" for a job? And look at how many "web 2.0" style companies we have had go tits up in the past couple of years, and with the economy like this I would expect to see plenty more. When a company is struggling, which you may not even have a clue about, the odds that they are going to spend the money to do best data backup practices is virtually nil. Then what happens when the server dies, or the drive that contained your data goes tits up? "Ooops, sorry about that" as most of these companies have it in their TOS that they pretty much ain't responsible for jack.

So no thanks. I can slap a cheap USB drive and back up my data anytime. I even have my data partitioned to where I can back up all the important stuff to DVD without having to back up the whole OS. I can encrypt the backups so nobody can use them but me, slap the USB or DVDs in my laptop bag and have them wherever I go, even if the place I end up at has "10k on a good day" dialup, which you'd be surprised how many places in the USA are still like that. So I think I'll just stick with what works no matter how good or poor a net connection I have tomorrow is, or without having to worry about whether or not the cloud company is doing well financially or using best backup and privacy practices. Call me weird but I like being in control of my data, thanks ever so much anyway.

Re:Take away the cloud

[RudeIota]

There is nothing amazing about applications in a browser, it is not necessary, and while it is convenient at times (at a computer that is not your own), when available a native code app will usually do the same job but "better".

I think you already know this, but it's not about doing it better: It's about ubiquity.

While I agree it's another "layer of abstraction", the point of using a browser is access from anywhere / anything. Whether it's a desktop, netbook, phone, Linux, Windows, Mac OS X, CISC, RISC etc.. you can potentially run whatever it is and have the same exact same experience on any platform. That's pretty cool.

By using native applications, you just don't get that ubiquity. People also want stuff that just works... Users don't want to install different versions of the same app on different devices and deal with the occasional issues of not having the correct libraries, UI differences and so on. It's also a terrible pain for the developers: Who wants to maintain, port and compile dozens of versions of the same product?

With the idea of native apps not being practical for the purpose of ubiquity, one would probably point to cross-platform frameworks, like Java or GTK. That's fine and good, but these things require some "coaxing", if you will, especially in the UI department. A really simple app might be work just fine, but you have to be careful about using OS-specific functions and more complex programs sometimes need to be changed substantially. Applications that use web browser technologies don't really suffer from this.

So, how do we keep the experience exactly the same on every system, every device, everywhere? Why not use well supported web technologies like HTML, CSS, Ajax, Javascript, Flash etc? To me, it just seems like a natural extension of that greater desire to be "ubiquitous". There are certainly limitations and a native app will be faster, but is it better? I think that depends on how you define "better". Functionality and compatibility are the main concern of web apps and to address these things, using the browser as a platform of sorts... well... it makes a lot of sense.

Re:Take away the cloud

[moosesocks]

Google have already started to tackle this one.

Although it's got to be difficult to mirror all of the application logic offline, the Gears apps that I've used thus far make a valiant attempt, and seem to preserve the core functionality.

Re:Take away the cloud

[vtcodger]

***You try, getting three different clients working against a database from the same vendor working properly. They all crave different versions of dotnet, java or whatnot and any new version of the client software demands countless hours of testing just about every possible combination of apps.***

Thanks, no. Been there. Done That. You're right. It is a nightmare.

But I'm curious why you think, as you apparently do, that switching to "the cloud" is going to be better. From where I sit, "the cloud" looks like a huge glob of poison gas. More standards than anyone can keep track of. But no one complies with them anyway. No discipline. Very little common sense. I suspect where the cloud is headed is a worse shambles than the current desktop plus latency and bandwidth problems. And security ... what security? Do people seriously think that "Never run as root" is going to prevent the ongoing security disaster?

Fortunately, I am retired and no longer have to make a living fighting with computers. I have a lot of sympathy for those who are not as lucky. Fasten those seatbelts folks, the next couple of decades are going to be one bumpy ride.

Re:Take away the cloud

[rtfa-troll]

You are of course right in theory; But this is a typical case that "in theory theory is the same as practice; in practice things are different". HTML has somehow managed to get the right balance to be much better than other applications. Primarily, there are no viruses written in HTML and HTML+Java(ECMA)script has almost no practical viruses.

The key advantages of HTML / ECMAscript / HTTP include

• not blocked at the corporate firewall
• has a subset which is pure data and easy to be sure is safe
• has a subset which is compatible across many different platforms for many different years
• is not controlled directly by a company with criminal tendancies

Every other option has serious drawbacks

• Java / .Net - too heavy; the minimal application requires loads of extra stuff
• Java / .Net / C++ - non trivial to package.
• C++ / .exe - too much history with trojans / too much incompatibility e.g. try developing one for Windows 200 working on Vista; compare with
• Anything which doesn't go over ports 80 or 443 - blocked by the firewall
• Anything containing executable content - blocked by the corporate mail filter
• .Net stuff - doesn't run on out of the box Ubuntu or Macintosh / not cross platform.

Disclosure: I'm currently interning at MS.

your honesty is appreciated. When you are just starting in the job market, any good job seems like a good idea. Please remember you have years and years of work, ahead. Taking ethical choices is a seriously good idea. When your CEO is threatening your president with firing you then you seriously should consider if that's a company you want to work for.

Re:Take away the cloud

[blowdart]

Funnily enough Microsoft office already has this, with Office Live. I have Open from Office Live and Save to Office Live in my file menu.

Re:Take away the cloud

[Jurily]

As far as syncing, there is nothing stopping native apps from syncing to "the cloud".

Except common sense, of course. I, for one, tell the new corporate overlords to stay out of my computer.

Re:Take away the cloud

[Locutus]

native code might sometimes be optimal but guess what? A heck of a lot is good enough in a browser and that means it runs on nearly every device and not just the ones with the Windows logo on them. Look what happened to the netbooks when Windows got ahold of them. They got fat, they got shorter battery lives and they got more expensive. The OLPC XO is another example, Windows doesn't fit on everything but the browser does. Good enough.
 

LoB
 

Re:Take away the cloud

[Locutus]

that's why Microsoft does not like it. They must play here but like with MS-OOXML, they will do anything and everything they can to somehow tie it to Windows, to one software environment. After all, their profits and growth come from Windows and without Windows, they house falls down.
 

LoB
 

Re:Take away the cloud

[drx]

Dude the main thing is that people click on an icon that shows a diskette!

Re:Take away the cloud

[dna_(c)(tm)(r)]

With the idea of native apps not being practical for the purpose of ubiquity, one would probably point to cross-platform frameworks, like Java or GTK. That's fine and good, but these things require some "coaxing", if you will, especially in the UI department. A really simple app might be work just fine, but you have to be careful about using OS-specific functions and more complex programs sometimes need to be changed substantially. Applications that use web browser technologies don't really suffer from this.

Sigh, it happens too often that people say "you have this problem with (C|C++|C#|GTK|Qt|$YOUR_FAVORITE_LANGUAGE)+ and Java".

Java has webstart since 1.4 (2001) which uses JNLP (Java Network Launching Protocol ). From the desktop user's point of view.

WS automatically downloads new versions of the software, checks java versions etc. If you want to see it work and you have Java installed (>=1.4) you can try these Java 3D examples to see how the experience is.

Re:Take away the cloud

[El_Muerte_TDS]

Indeed, "Save to the cloud" should be renamed to "Vaporize".

Re:Take away the cloud

[Anonymous Coward]

You don't really think local internal storage is more reliable than the cloud?

Most users really don't have raids nor proper backup. I feel my documents are much more secure (and accessible) in my gmail than on my disk partition (come to count it, currently I have 9! not counting numerous USB disks or flash cards). And each can die moment, and I don't even think to start taking care of backuping them all up. Even if they don't die, get rewritten or formatted I will add or replace another computing device and migrating is always a pain, even for me.

Number of PCs per user is only going up, and the cloud is the only solution to the data mess that comes with.
And with smartphones and blackberries things get even better if you opt for the cloud.

I agree speed does not seem to be an issue (I still waste more time locating / browsing for my documents than it takes me to move them across the net).

Internet connection is reliable as electricity or mobile networks (if not more so :-) and we are so depended on it anyway (probably more than we are aware) that it does not make sense to constrain and lock down your documents with no clear benefit.

All my data on disks, diskettes, cdroms, printouts from 10 years are lost or destroyed except for my online home space at my ex college and my geocities page (oh wait, that's gone too :).

Re:Take away the cloud

[mcvos]

People say this a lot, but it mystifies me. The cloud is empirically much more reliable than internal storage; hard drives crash all the time and lose *all* their data. Unless you're running a RAID and doing daily offsite backups your data is safer in the cloud because they do it for you.

What does RAID have to do with it? Regular backups have always been important, and my internet connection drops more often than my harddrive. Local storage means you're not dependent on internet connections, online storage means you can more easily access it from different locations. That's what the trade-off is here.

Nobody I know has ever lost any data stored on GMail, Flickr, or similar. The worst that I've ever seen happen is someone might not be able to log in for a few hours; maybe up to a day in an extreme case.

And that can be a big problem if you've got a big company working on something important. I've seen companies twiddle their thumbs all day because internet was down. Making yourself even more dependent on that doesn't sound like a step towards reliability.

On the other hand, practically everyone I know has experienced a hard drive crash, sometimes losing valuable data forever,

Then they should have backed up their data.

Re:Take away the cloud

[CoccoBill]

What I think would be best for Google would be to fork a version of OOo to include "Save to the cloud" support and integration with Google Docs.

I'm sure some home users might find that useful, but good luck selling that to businesses.

"Now, where's that option to save this classified document somewhere on the internets?"

I'm hesitant to save even any of my personal stuff to google's services, a business that'd allow it will not be in business for long.

Re:Take away the cloud

[Mystra_x64]

They're compatible with all operating systems, portable, simple to build, and really really light weight.

Yeah, right until "Your browser is not supported."

Re:Take away the cloud

[digitalchinky]

Lots of businesses do this already - sometimes it's just more convenient to use a drop in solution rather than roll out your own, you weigh up the risks first obviously.

Netsuite is just one of several that cost anywhere between 5 and 6 figures to set up, and they don't exactly lack clients.

Re:Take away the cloud

[diegocgteleline.es]

I think it will be a long time before the internet/cloud can compete with local internal storage.

Yeah, tell that to the email clients...

Re:Take away the cloud

[arethuza]

Also, Office applications have been WebDAV clients for years - they will save stuff to a remote server fairly happily.

Re:Take away the cloud

[LKM]

THe browser offers a number of advantages to app developers. Simple deployment of new versions, built-in strong encryption, a simple API for server connectivity, and finally, it's pretty much platform independent, anything from an iPhone to a Windows PC will be able to run your app if you're reasonably careful when developing it.

Re:Take away the cloud

[Corporate Troll]

my internet connection drops more often than my harddrive.

While true that harddisks fail rarely (except from pretty much all 160Gig HDs I have owned, even from different brands), a failed harddisk will bring you down hard. An internet connection that goes down? That's as much work as reconnecting.... Many modern routers even do that for you if they sense a disconnection.

My Internet is down maximum 2 times a year, and I have a old (from 2003) ADSL Modem with a OpenBSD router behind it. Restarting the Internet is as much work as running a script which I called "restartdsl". (Kills existing connection, puts tun0 down, flushes routes and reconnects... easy peasy)

Try fixing a failing harddisk that way....

Re:Take away the cloud

[Corporate Troll]

Actually even my 67 year old clueless dad has a USB external with one touch backup.

I would argue that your dad is not clueless. Do you know why he does this? I know: He got burned before and that's how he learned he got to make backups. Since he learned he is by definition not clueless

Give your old man some credit, okay?

Oh, and as for a final remark... People have only started doing this the last few years (only those that got burned before, mind you) Before ubiquitous cheap USB harddisks, backup simply was too expensive for the average home user. (I'm someone who backed up on Iomega Dittos... don't ask... it was not cheap, but cheap enough, but a fucking pain!)

Re:Take away the cloud

[mcvos]

While true that harddisks fail rarely (except from pretty much all 160Gig HDs I have owned, even from different brands), a failed harddisk will bring you down hard. An internet connection that goes down? That's as much work as reconnecting.... Many modern routers even do that for you if they sense a disconnection.

I'm not talking about a dialup line that's dropped. I'm talking about a corporate ISP that's crap.

If your internet connection is gone for a day, and all your documents are only stored online, then the entire company can't access the documents for a day. If the harddisk the documents are on crashes, you can restore them from backup, which usually takes quite a bit less than a day. And it also happens less often. Unless you have a redundant internet connection, but if that's the case, your file server is also bound to have some redundancy.

If you want to be able to access your files from different locations, then of course an online backup makes a lot of sense. But you should never rely on only online if you want reliable local access to your files.

Re:Take away the cloud

[Corporate Troll]

I'm not talking about a dialup line that's dropped. I'm talking about a corporate ISP that's crap.

Don't you think the problem is somewhere else? Like having a crap corporate ISP? Switch! It's that simple...

Apart from that, in a corporate setup, there should not be local storage at all. "My Documents" should be pointing to a network accessible share and those should have nightly backups and offsite backups. Technically they pretty much already have their "cloud", even if it is implemented in a different way.

I was indeed thinking more of home users (dial-up, are you kidding me? Nobody uses dialup anymore... not on my continent). Corporations require other solutions, one of which I described above. I'd expect only small and medium businesses to rely on services on the Internet and believe me, those do not have the infrastructure and backup facilities Corporations have. (Also, they are mostly on ADSL, which is restartable as I said... For them, it means rebooting the router, but that's not hard)

Still, doesn't change the fact a Corporation having a bad ISP should take action and not sit there like a lame duck. Actually, I'd blame the IT guys of that corporation for failing to provide adequate service to the company.... Sorry, that's just the way it is.

Re:Take away the cloud

[Kashgarinn]

It's all about different platforms, and how the lowest platform kinda works on all other higher ones.

Creating (office) tools for a browser engine to be able to display means that you can use those tools on all platforms which have access to that engine. You're no longer talking about PCs here, you're talking about phones, home entertainment systems, other OSes, and probably standalone TVs, basically any platform which has a screen and an ability for user interaction.

If you create a tool which works in a browser, you open up your tool to all of these platforms, and if you create the tool to work well on the lowest platform, the user experience will always be good on any higher platform.

the tool might not be optimized for a x86 processor, but if it's optimized for an even lower platform, it should run well in a browser on a PC.

- I'm not surprised that someone at microsoft doesn't get that.

Re:Take away the cloud

[TheLink]

> Java has webstart since 1.4 which uses JNLP...

1) From a firefox user's PoV - it looks like "Opening a Program from a Website" which is what everyone keeps yelling at them NOT to do.

2) It took me more than 5 minutes to download a simple program that just draws 3D spheres (from your link). Yeah I have a crappy connection, but I doubt "corporate" java apps are going to be small (and I've seen them being updated every few weeks - which means everyone has to redownload). They're fine over the LAN, but this Cloud thing...

FWIW, I've tried the 4K java game stuff and many are great (and download quickly), but there are very few java programmers who can and will do that for "office/corporate apps".

Re:Take away the cloud

[Computershack]

On the other hand, always-on internet access is real for some people already.

And as people who use Gmail and Google Apps for business have found out, just because you can connect to the internet doesn't mean that the cloud service is going to be up.

Re:Take away the cloud

[YttriumOxide]

True story though - I was showing a member of the younger generation a floppy disk that I had lying around, and the first thing he said was, "ah, it looks like the 'save' icon!". To many of the younger generation, the icon is an abstract concept meaning "save", not a representation of a disk.

Re:Take away the cloud

[Anonymous Coward]

You are wrong. A cloud is nothing more than a mass of vapour and cristals floating in the air. The computer "cloud" is nothing more than a marketing term that refers to an external server. It may be a single server and it may be a huge cluster but you always interface with it as if it was a single server. It's like saving info to a file in your computer. You don't need to know how to draw a precise system diagram in order to save the file in your system. You may save it to your HD, to a RAM drive, to an external HD or even in a remote computer in some other continent through some remote shell utility but you are still saving your data to a file. Trying to make believe that uploading your info to an external server something magical and intangible as "the cloud" is embracing marketing stupidity just for the sake of it. So please don't follow on that tendency, as it only makes you participate in the dissemination of idiocy and as a consequence look bad.

Re:Take away the cloud

[peragrin]

The cloud is the new THIN client. The browser is the new OS for said thin client.

while different browsers do implement things differently, there is a standards running the whole show and an approval process. previous thin clients were proprietary in one form or another.

Re:Take away the cloud

[mcvos]

Don't you think the problem is somewhere else? Like having a crap corporate ISP?

Maybe, but it still happens. Servers can go down, even big ones. Even gmail has had its problems. Relying on 100% uptime from outsiders is an invitation to disaster.

Apart from that, in a corporate setup, there should not be local storage at all. "My Documents" should be pointing to a network accessible share and those should have nightly backups and offsite backups.

That is exactly what I'm saying, and it's not the same thing as storing it on a third party server over a fourth party internet connection.

I was indeed thinking more of home users (dial-up, are you kidding me? Nobody uses dialup anymore... not on my continent).

You gave me a different impression with your comment about dropping internet connections that could quickly be restored. My experience with dropped internet connections is that usually something broke on the ISP side. Anything less than that is just not terribly serious.

I'd expect only small and medium businesses to rely on services on the Internet and believe me, those do not have the infrastructure and backup facilities Corporations have.

You make it sound like backups are really complex. They're not. Small companies can and should have adequate backup facilities.

(Also, they are mostly on ADSL, which is restartable as I said... For them, it means rebooting the router, but that's not hard)

That still doesn't help you if it's a problem on the ISP's side or even further upstream.

Still, doesn't change the fact a Corporation having a bad ISP should take action and not sit there like a lame duck. Actually, I'd blame the IT guys of that corporation for failing to provide adequate service to the company.... Sorry, that's just the way it is.

You mean "anything as long as you don't have to make backups"? Sorry, but that's just ignorant. It's very easy to take your own responsibility here and make you less dependent on others.

Re:Take away the cloud

[tclgeek]

Surely you must have meant to say "... huge step forward for admins".

What many (most?) application developers forget is that applications exist to make end users more productive, not admins, and there are precious few web based apps that are better than their desktop counterparts.

Re:Take away the cloud

[miffo.swe]

The single most frustrating thing for a user is when something doesnt work wich in a normal corporate setting is pretty much all the time. Its very expensive to keep things running.

If you have had bad experinces with web applications chances are you have been running ActiveX kind of "webapps" wich is the worst of both worlds.

Re:Take away the cloud

[larry bagina]

From where I sit, "the cloud" looks like a huge glob of poison gas.

That's not the cloud... cowboy neal had chili for dinner last night :(

Re:Take away the cloud

[idigitallDotCom]

What is with all of you suggesting "the cloud - the cloud - the cloud" all the time. I don't trust you. Why should I? Why should I trust my colleague who wants to borrow my 16Gb flash drive because his doesn't have space, with my "classified" company information that I store on it? And now everyone's suggesting "the cloud - the cloud - the cloud" with GoogleDocs and OpenIDs.

The cloud is NOT secure. Heck, not even passworded PDFs and DOCs are secure - forget about uploading it onto someone ELSES server and *hoping* that one nosy-parker low level administrator who doesn't have enough to do isn't snooping where he shouldn't be.

I need proof that the cloud is secure before I'll upload "secret" files to it. I'd much rather password protect the document, then zip it up two or three times in a passworded zip file, then hide it in the hidden folder on my micro SD card, and then embed it under my little fingernail before I trust that frikken' cloud!

Re:Take away the cloud

[Colonel Korn]

Getting rid of stupid clients would be godsent for any admin in the world. Having all applications in the browser would be a huge step forward.

You try, getting three different clients working against a database from the same vendor working properly. They all crave different versions of dotnet, java or whatnot and any new version of the client software demands countless hours of testing just about every possible combination of apps. Upgrades are pure nightmare. Couple this with locked down desktops, profiles that has to be managed and policies that needs hard testing before you alter a single setting.

Getting rid of all those problems alone would be worth serious money for any company. Added benefit would be that backend services would be totally decoupled from what OS the client runs. Microsoft will fight this for all they are worth.

Getting rid of stupid houses would be a godsend for any contractor in the world. Having all people inside a mud hut would be a huge step forward.

You try getting three different walls working against the same roof from the same floor working properly. They all crave different stud layout, nails or whatnot and any new wall demands countless hours of planning for just about every possible combination of other walls. Electrical is a pure nightmare. Couple this with locked doors, surfaces that need to be painted and building codes that need hard testing before you can finish a single room.

Getting rid of all those problems alone would be worth serious money for any construction company. Added benefit would be that the extension cord used to steal electricity from neigbors would be totally decoupled from the type of mud hut. Legacy housebuilders will fight this for all they are worth.

Re:Take away the cloud

[krewemaynard]

But Microsoft people have a good point about the cloud. Forget speed, think about reliability. And by reliability of the cloud, I actually mean reliability of your internet connection. I think it will be a long time before the internet/cloud can compete with local internal storage. So for Google to compete, cloud features are an awesome additional feature, but to really succeed, I think they need to be able to go toe-to-toe with Microsoft on the desktop without requiring an internet connection.

Google Gears pretty much takes care of the flaky connection problem. Keep a local cache, and sync everything when the Net connection comes back up.

I would love to see Google Docs incorporated into OOo. I've tried an OOo plugin before that was supposed to save to and open from Google Docs...don't remember the name of it, but it kinda sucked anyway. Some enterprising Google engineer could probably whip up a solution in their 20% time, but I'd really love to see it as an official project.

Re:Take away the cloud

[sorak]

I use four different computers on a weekly basis, and on one of them, I cannot install software. So, it's not just about platform. It's also about being able to go to any random "box" and access your documents.

Of course I should probably use Google docs more, but my night job is teaching Office 2007, which means having to get used to all the changes they made since XP.

Integration plugins would be nice, however. I am currently using ftp to sync the different systems, and would like something simpler.

Re:Take away the cloud

[docbrody]

You don't really think local internal storage is more reliable than the cloud?

Yes I do... Well, as I said in my original post ''And by reliability of the cloud, I actually mean reliability of your internet connection''

I am not really saying anything new or different. I am not talking about backups, or long term retention of data, or any of that. My original post was just an acknowledgment that to compete with Microsoft, Google has to have an offline strategy.

There are still plenty of times when I can't access the net: In planes, trains and automobiles. At my parents house. In pretty much all of Vermont. Even at the local Starbucks in the west village in New York Fucking City of all places (they want me to pay for wifi). And very very very rarely my internet goes down at home.

So its not even about unreliable ISPs. Its just that I move around a lot. On one level its not fair to call these 'reliability' issues at all. But on another level, from the perspective of the average Joe, if he can't reliably access his data, then that's a hassle and gives local internal storage a perceived reliability advantage.

Re:Take away the cloud

[docbrody]

Google Gears pretty much takes care of the flaky connection problem

When you can take 'pretty much' out of that sentence, then we are in business.

How does it handle situations like when you are on a plane and and you want to pull up a year old document that you could use as a template for the one you are working on now?

Re:Take away the cloud

[docbrody]

Yeah, tell that to the email clients...

I do in fact tell that to my email client regularly. It doesn't seem to listen though. The damn thing still refuses show me my new mail until my plane lands in Newark.

Re:Take away the cloud

[drinkypoo]

So no thanks. [...] Call me weird but I like being in control of my data, thanks ever so much anyway.

Explain pls why you have a gmail address, then. You don't even know what you're saying.

You complain about gmail being offline, but the truth is that everyone has downtime. Meanwhile, gmail offers Offline Gmail via Google Gears; Even while offline you can view your downloaded mail. Ignorance? Disingenuousness? Because the simple truth is that an ISP can have POP/IMAP server failure, or lose one of their couple of connections. Meanwhile, Google (and other providers of "cloud services") are gaining experience in running their distributed systems and becoming more reliable.

It's clear that retaining your own backup copy is always a good strategy, but history shows that these copies can be lost. It's also clear that letting control of your data out of your hands is dangerous, but that's why we have encryption. Having a local backup and an offsite backup is an absolute necessity for any data which you cannot stand to lose. I have probably terabytes of data and if I lost it in a fire I might cry about it, but no lives would be lost, scientific progress would not be harmed, et cetera. The greatest tragedy would be the pollution produced by allowing the stuff to burn, so yes, I work pretty much on the one-backup system, and a lot of my data has no backup at all... but nobody cares that much, not even me. What I do care about is keeping my email, and google makes that trivial. In addition, if I want [relatively] secure communications I can use FireGPG and disable autosave of drafts. The situation with storing the body of your data "in the cloud" is pretty much identical, except you have to deal with bandwidth issues as well.

Re:Take away the cloud

[Corporate Troll]

My experience with dropped internet connections is that usually something broke on the ISP side.

Mine is that it rarely is on the ISP side. In the 6 years (wow, 6 years already!) I have had exactly one "drop" of connectivity which was the ISPs fault. That's not bad for residential ADSL. Companies should demand better uptimes than that.

You make it sound like backups are really complex.

No, I did not.... However, I do know that small and medium businesses usually don't have inhouse IT people and actually don't know how to do it. Don't forget that in many SMBs, it's Secretary Sallys who knows Word that does all IT. No, I'm not kidding.

You mean "anything as long as you don't have to make backups"?

No, that's not what I said at all. Large corporations should have their data in-house and have offsite backup. However, having an ISP that fucks up days of connectivity is a no-no. At that point you search another IT crew who actually negotiates correctly with ISPs. Meaning: give them hell if the Internet drops a whole day.

Re:Take away the cloud

[blueZ3]

But it's going to be a trade off (like everything else in life) right? Sure, you'll be able to access you content in the same format using the same tool from anywhere... as long as your Internet connection is live. What you're trading off for ubiquity is some level of availability.

As an example, we keep our budget in an Excel workbook that my wife cooked up. The budget has categories with "allocated" monthly amounts and we put in every expense and then check the spreadsheet against our credit card and bank statements at the end of the month. It would be cool if I could access this spreadsheet from anywhere--say when I'm on vacation to access the "original" version rather than moving a copy onto my laptop. But it would be very uncool if I wanted to put expenses in but couldn't access it because AT&T was on the fritz again.

Where I live, the Internet is much less reliable than the electricity. Not that it goes out all that often, but in the four years I've owner my house there have been maybe three power outages--and maybe 10 or 12 times that our Internet connection has either stopped working or been seriously degraded.

Then there are security issues. Would you be willing to upload a document that contains your SSN or CC number to Google docs? What about to some other on-line application provider? And if not, where do you draw the line... at your name and address? Personal details and family photos? Because once you draw a line, you're saying that for some content you'll use the online application and storage, but for others you'll use a local app and storage--which makes things more complicated.

In addition, I have concerns about longevity. Sure, Google is the new hotness right now, but what happens to all my stuff when big G goes belly up in 2050? Will I have enough advanced warning to download all my content to local storage?

Sorry for the long post--I'm not arguing that this isn't something that would be without usefulness, I'm just pointing out that there are trade-offs to be made

Re:Take away the cloud

[Sylver Dragon]

Do you really think it's going to be that different? Right now "cloud computing" is in that wonderful conceptual stage where everything is puppy dogs and roses. Unfortunately, those puppy dogs will eventually grow up and chew up the furniture and shit everywhere; and the roses will die and the thorns are going to stick you when you try to fix them. Every technology which comes along is supposed to make things easier, better, more secure and do the dishes. None of it does. Sure, it may be better than what we have now, then again, maybe not.

You try, getting three different clients working against a database from the same vendor working properly. They all crave different versions of dotnet, java or whatnot and any new version of the client software demands countless hours of testing just about every possible combination of apps.

So, instead I get three different browsers, inside each of which I am supposed to run three different versions of a database client, and probably against a database I no longer have any real control of. Oh, and some jerk-off somewhere in the organization probably wants to run the whole thing on a different OS, because you will still need some sort of local OS to boot the damn thing. Congratulations, you've now added yet another layer of complexity to fail and troubleshoot. Thank you, no.

If anything, what you are after is virtualized applications, the previous puppy-dogs and roses technology. At the very least you, as the admin, still have control of the system end to end, and it really does decouple the applications from the instance of the OS. It still has it's own host of issues, such as, "I need .NET 2.0 with this app, .NET 3.0 with this app, and Will's API Really Good At Running Bob's Lines (WARGARBL) on this other app." which can make it fun. Of course, it is yet another layer of complexity to fail and troubleshoot, but at least I can look at the whole server-client setup.

Upgrades are pure nightmare. Couple this with locked down desktops, profiles that has to be managed and policies that needs hard testing before you alter a single setting.

If you're worried enough about security that you go through unit testing for every profile change, why in the world would you trust your data to an untrusted network? It's one thing to run an internal terminal server with thin clients, but that isn't what Google, Microsoft, et al. are after. They want to run the terminal server, file server, and everything else for you. And all of your work, data and information will be secured on the internet through an encrypted pipe. Sure, it'll probably keep out the third party attacks, right until someone at Microsoft or Google decides to fuck everyone. And, of course, the contract you will sign for their services will indemnify them against lawsuits. And they'll probably still foist onto you the job of creating your own profiles, which will still need just as much work to secure.

Getting rid of all those problems alone would be worth serious money for any company.

Yes it is, and it sounds like they are spending the money in the right place to make those problems go away. That spending is on you. If you care enough to do the work you have outlined above as an admin, and are still looking for ways to make it better, then you are earning your paycheck.

Re:Take away the cloud

[krewemaynard]

How does it handle situations like when you are on a plane and and you want to pull up a year old document that you could use as a template for the one you are working on now?

Same way Offline Gmail works right now. It downloads a local copy to your PC and keeps it synced with the cloud while you're connected. When your connection's gone, you work as usual. Once you re-connect, everything is synced again in the background. It works well.

Re:Take away the cloud

[hairyfeet]

Uuuhhhhh.....The Gmail address is a spam dump. Why would I actually care if Google has my spam or not? They are welcome to all my spam, anyplace, anytime! For my real mail i have an address with my ISP that not only has a web interface but works real nice with Thunderbird even on a shitty dialup connection. Have you ever tried Gmail over dialup? It ain't very pleasant.

Look, I don't have anything against someone deciding a cloud service was right for them. just as I don't have anything against those that decide a thin client/server combo is the right way to run their business. I just said that when you add in reliability issues, privacy issues, the TOS that basically lets them off the hook no matter what goes wrong, etc, it isn't right for me and my customers. For my customers i sell them a little 2.5in USB HDD with one touch backup. At the end of the day they push the button and once it is done they slide it into their briefcase and go home. It makes offsite backups easy without having to worry about the data.

But where I live now, as well as where I have lived in the past reliable Internet can sometimes be nonexistent. no matter how good Google gears or any of those other cloud services are, you still need a connection to make it really function correctly . It is kinda like how everyone is talking about how streaming everything is the future. Maybe, but when they are places in downtown Nashville where i could not get anything other than dialup because the local duopoly didn't run there it certainly isn't the present. maybe when like Korea we have nationwide broadband things will be different. i was simply pointing out we still have quite a ways to go for cloud computing to be more than a big city service here in the USA.

Re:Take away the cloud

[VGPowerlord]

Getting rid of browser clients would be godsent for any admin in the world. Having all applications as a native application would be a huge step forward.

You try, getting three different web browsers working against a site using the same HTML/CSS/Javascript working properly. They all crave different versions of styles, scripts, tags or whatnot and any new version of any of the major browsers demands countless hours of testing for just about every possible combination of apps. Upgrades are pure nightmare. Couple this with locked down desktops with only Internet Explorer 6 in corporate settings, graceful degradation if scripts are disabled and usability requirement laws that need hard testing before you alter a single setting.

Getting rid of all those problems alone would be worth serious money for any company. Added benefit would be that backend services could be totally decoupled from what OS the client runs. Google will fight this for all they are worth.

Re:Take away the cloud

[shutdown -p now]

Primarily, there are no viruses written in HTML and HTML+Java(ECMA)script has almost no practical viruses.

Simply because pure HTML/JS applications are very strongly sandboxed - because standard APIs don't provide any way to access local filesystem etc. Something you can't really write a working virus or worm without.

But this will change with HTML5, so...

Java / .Net - too heavy; the minimal application requires loads of extra stuff

Only if neither of runtimes are preinstalled on the client box. Sun boasts of JRE market penetration on the desktop; Microsoft doesn't, but Vista comes with .NET 3.0 out of the box, and Win7 will come with .NET 3.5 (and also all XP/Vista boxes which had .NET 2.0+ installed on them are upgraded to 3.5 via Windows Update)

.Net stuff - doesn't run on out of the box Ubuntu or Macintosh / not cross platform.

Depends on which subset we're talking about. Silverlight does run on Macs, for example.

Re:Take away the cloud

[docbrody]

Fair enough.

Re:Take away the cloud

[BeanThere]

Having all applications in the browser would be a huge step forward.

Urgh. HTML is the crappiest "platform" ever. It was never designed to be an application platform, it's just 'de facto' evolved into one, and evolved in a grotesque, messy way at that. I think making the Web browser a "platform" was the biggest step backwards in the history of software development ever, we've been set back years if not decades in terms of app development speed, capabilities, robustness, performance, size, requirements, bloat, etc. ... something brand new and cleanly designed for the purpose would've been much better. Heck, X would been much better, and was already "cloud-friendly" and generally "buzzword-friendly" anyway.

Re:Take away the cloud

[BeanThere]

While I agree it's another "layer of abstraction", the point of using a browser is access from anywhere / anything. Whether it's a desktop, netbook, phone, Linux, Windows, Mac OS X, CISC, RISC etc.. you can potentially run whatever it is and have the same exact same experience on any platform. That's pretty cool.

Sure the idea is cool, but the implementation is anything but. And there've been other much better implementations of the same idea; they just haven't been able to gain the same traction; mainly for historical reasons the web browser has evolved into the 'cross-platform application environment of choice', not because it was technically a great solution to the problem. It's not.

Re:Take away the cloud

[BeanThere]

I use four different computers on a weekly basis, and on one of them, I cannot install software. So, it's not just about platform. It's also about being able to go to any random "box" and access your documents.

Oh, you mean like X / UNIX were designed to do and able to do decades ago already - elegantly, securely, fast, and bloat-free.

How many times is this industry going to re-invent the same wheels badly.

Re:Take away the cloud

[BeanThere]

As far as syncing, there is nothing stopping native apps from syncing to "the cloud".

Our customers are saving their most important and crucial intellectual property in our database applications - their 'crown jewels' - is it really a great idea to just randomly entrust saving your company's most precious IP (which, if it leaked, could sink your company) to "the cloud", an amorphous entity consisting of computers that could be anywhere, and could be accessed by who-knows-who amongst varied and changing subcontractors and employees, and may be susceptible to hacking or poorly scrupled individuals e.g. within subcontractors who have access to bits and pieces of this network?

Re:Take away the cloud

[MikeBabcock]

What is it with you people? They're applications whether they run in a browser or not.

An AJAX website is an application. GMAIL is an application. It just so happens that the GUI is rendered by a second application (your browser). This is no different than switching GUI toolkits when making a traditional app.

All you're doing is using a browser + html + javascript as the language and gui toolkit. This is not new, its not exciting and its not even interesting.

If you think that a browser + html + javascript is a better fit to your application than .NET, great, do it that way. If you think Python + Gtk is better, then do it that way instead. Java +AWT might be the best choice, who knows.

This crazy 'browser replacing the OS' stuff is dumb, the browser has simply become a programming environment much like Java was before it, and traditional language+toolkit options were before that. It just so happens that the web browser became pervasive in a way that Java didn't and so its the most readily available toolkit for many programs. That doesn't make it implicitly better than anything.

From a security and debugging standpoint, I personally think browser based applications are a kind of security hell on earth. I have to trust a huge monolithic code base I often can't see (IE8, iPhone browser, etc.) to properly interpret my intentions in a way that I can more completely control in a traditional environment. Many see this as a good trade-off, some of us see it as horrifying.

Facebook is a good example. Its a perfect fit to the browser as an application interface paradigm, and yet people are constantly surprised by new and interesting ways their private data is getting away from them because of how the interface hides reality. It will happen with the next big thing too.

Re:Take away the cloud

[MikeBabcock]

Corporate Java apps would be served at LAN speeds of the corporate Intranet, and not limited by your Internet speeds.

For example, you might browse to an internal website that links to 10.11.12.13 and launches software hosted on 10.11.12.17 using the database at 10.11.12.23, all of which are 100Mbit or gigabit+ connections to your PC.

As far as the user's concerned, the app launches near instantly, and updates itself automatically. Its very very nice. As a non-Java programmer, its a feature I'd love to have worked well into Python application distribution.

Re:Take away the cloud

[MikeBabcock]

You do know that the browser which runs your HTML app itself falls under the category of "Every other option ..." right?

The browser itself which you take as implicit is just another application. Browsers have trojan/virus/compatibility problems. Browsers have pop-up blocking and other firewall software issues. Browsers are susceptible to bugs like the cross-site scripting ones we've seen and others that are worse. Browsers can act differently on different platforms.

The benefits don't exist, they're just abstracted in your mind because you ignore what's under the layer of abstraction.

Re:Take away the cloud

[mcvos]

Mine is that it rarely is on the ISP side. In the 6 years (wow, 6 years already!) I have had exactly one "drop" of connectivity which was the ISPs fault. That's not bad for residential ADSL. Companies should demand better uptimes than that.

My home ADSL is doing quite well too, but the cable I had before that was crap. At my previous job, our corporate DSL dropped so often that I insisted we get a backup internet connection, and I believe my current job has also experienced a hiccup in the half year that I work there. And these were both web companies, so we really can't do anything when the net is down.

However, having an ISP that fucks up days of connectivity is a no-no. At that point you search another IT crew who actually negotiates correctly with ISPs. Meaning: give them hell if the Internet drops a whole day.

Don't worry, the ISPs get hell. But you still lose a day. Shit just happens, and the only way to prevent it is redundancy. You've clearly ben lucky, but not everybody is that lucky. And blindly counting on luck is risky.

Keep in mind that 99.9% uptime still means it can be down for half a workday every year. No ISP can guarantee perfect uninterrupted service. You can get lucky, or you can get unlucky. Or you can get a redundant connection, but that's not something your small business with Secretary Sally is going to do.

Re:Take away the cloud

[sorak]

How many times is this industry going to re-invent the same wheels badly.

So your suggestion is that google set up a X server and just let everybody connect and use their openOffice installation?

It's only reinventing the wheel if they can do the exact same thing with existing technology.

Re:Take away the cloud

[rumith]

Yeah, right until "Your browser is not supported."

--
Quick way to get 30% Funny 70% Troll: defend Opera browser on /.

Should your comment be modded Funny or Troll?

Re:Take away the cloud

[Mystra_x64]

It's not always Opera who is not supported :)

Re:Take away the cloud

[badkarmadayaccount]

IOW, Web 2.0 = POSIX 3.0

Re:Take away the cloud

[badkarmadayaccount]

Repeat after me: Flash is not a true web technology. It is not an open implementation, or even standard. It was not meant for graceful degradation and easy interoperation through its lifecycle, unlike web technologies.

Re:Take away the cloud

[badkarmadayaccount]

The X server is on the side of the application user, the app is the client. And most X apps, their tool kits more precisely, aren't optimized and/or designed for higher latency / lower bandwidth connections. The X protocol was meant for LANs anyway. But yes, it would have been better to retrofit all that than reinventing everything bottom up, badly. Wasn't OOP invented with this sole purpose?

Re:Take away the cloud

[sorak]

The X server is on the side of the application user, the app is the client.

This would defeat the purpose of having one common platform available to all (or a very large majority of) users. Instead, it would require the end user to install X, and a specialized version of openOffice (or something similar). In this scenario, Google's server would be nothing more than a centralized file server, and the use of X would be unnecessary.

The X protocol was meant for LANs anyway. But yes, it would have been better to retrofit all that than reinventing everything bottom up, badly.

Considering your "The X server is on the side of the application user" comment, I am not quite sure how you would intend to make this work, but the X server paradigm seems to either work in one of two ways:

1. All the application processing is done on one server, which incurs a heavy load on this one server. The beauty of Google's approach is that they avoid the need for a cluster of servers that can host a million login sessions and a million instances of some office program simultaneously, by moving much of the processing to the individual user.
2. X is run on the local machine, and there is no ubiquity

Wasn't OOP invented with this sole purpose?

Yes, it was, but what you're suggesting is a massive redesign of an application to make it do something it was never intended to do. Sometimes it makes sense to clear the slate and start over with a design that better fits your purpose, even if it does involve "reinventing the wheel".

But with that having been said, there may be a way to fork X into a new version that pushes the processing onto the client PCs and produces better scalability. I don't know enough about x to tell you what difficulties this would entail (in essence, X would be like a virtual operating system), but it seems like an interesting idea for a advanced degree thesis or project, but it also seems like a little too much for Google to tackle for this one project.

Re:Take away the cloud

[rtfa-troll]

You do know that the browser which runs your HTML app itself falls under the category of "Every other option ..." right?

Absolutely... in theory. In real life the browser is, through various historical reasons, always there already.

The browser itself which you take as implicit is just another application. Browsers have trojan/virus/compatibility problems. Browsers have pop-up blocking and other firewall software issues. Browsers are susceptible to bugs like the cross-site scripting ones we've seen and others that are worse. Browsers can act differently on different platforms.

Somehow, you are right and yet, whilst I can easily point to several corporate desktops where the JVM is essentially disabled (my own included) I know no single one where there isn't a browser. In a "they're all turing complete" sense, I don't disagree with the great grandparent. However, if you are deciding which platform to deploy on; HTML is the one which will most easily just work. Platform to platform differences for basic old HTML 2 are trivial. Stick with the old and there will be no problem.

The benefits don't exist, they're just abstracted in your mind because you ignore what's under the layer of abstraction.

I ignore it because I can ignore it. The theoretical benefits do not exist. The practical ones are plain for all to see.

Developer candy?

[syousef]

Yeah developer candy is first on my list when looking at a product as an end user. I mean stuff security, reliability, etc. That's just all rubbish when I can make my developers even more diabetic.

Who comes up with this nonsense?

Re:Developer candy?

[AnalPerfume]

If the developers ain't interested in developing applications for it, the end user has no reason to install it.

Re:Developer candy?

[fabs64]

Fact is fella the end users don't decide on frameworks. And while sometimes decisions like that are made higher up often it is at the whim of a developer.

Re:Developer candy?

[syousef]

You think for commercial software they decide what framework? You think in any environment other than hobby projects that the primary selection criteria is developer candy???

Re:Developer candy?

[fabs64]

I am a commercial software developer.

Think about it. Who generally has the expertise and trust of management to make such decisions? If developers don't have the most input they certainly do have a say that holds influence.

'developer candy' can also be translated to 'lower barrier to entry' (cheaper programmers), 'faster ROI' (faster development for experienced programmers) and 'inherently higher quality' (larger cookie-cutter components)

What do you think developers enjoy working with? Inconsistent rickety unstable messes?

Re:Developer candy?

[syousef]

You're talking about hobby development.

Re:Developer candy?

[syousef]

What you're describing is not developer Candy. Developer Candy is a neat new scripting language or plugin that does something "cool".

What you're describing is:

- The Rapid Application Development paradigm which was too hastily abandoned.
- Code stability/reliability/predictability.

Re:Developer candy?

[fabs64]

HTML5 is a new scripting language? Or a plugin that does something "cool"?

That's user candy fella.

I pretty clearly was not describing a paradigm, I was simply pointing out that things developers like tend towards faster development leading towards faster roi.

Re:Developer candy?

[mcvos]

Not just that. Many developers prefer being paid for fun stuff rather than for boring stuff. It's certainly how I pick my jobs.

Also, hobby development has actually gotten quite big over the last decade and a half.

Re:Developer candy?

[syousef]

HTML5 is a new scripting language? Or a plugin that does something "cool"?

Misrepresenting what I was saying in order to ridicule it is just weak. You know perfectly well I was giving examples of what I consider developer candy and that I was not saying HTML5 is a scripting language.

That's user candy fella.

An Eclipse plugin is user candy?

I pretty clearly was not describing a paradigm, I was simply pointing out that things developers like tend towards faster development leading towards faster roi.

You're not clear about anything "fella".

Developers like lots of different things. Some developers love complex awful frameworks. See most J2EE frameworks.

Here's what Google needs

[bogaboga]

To give Microsoft something to seriously think about, Google needs an OS on the desktop. Android is a good start in my opinion. There are some efforts in this direction already. The good thing is that Android eschews X, which is a pain to work with in its current form.

Next, they will need [meaningful] applications that work no matter what platform one happens to be using.

Third, targeting Microsoft must not be the aim, it must be the unplanned outcome. The aim must be tp "please" we the users.

That way, Google will succeed on the desktop.

Re:Here's what Google needs

[moogsynth]

Their own OS? I don't think Google really needs one, but I agree it would be a step in the right direction. The main reason why Microsoft are so against openness in anything is that the fewer reasons there are to tie people to a specific OS, the fewer people will stick with that OS. We already have alternatives to Microsoft's offerings, so useful applications that work on anything are the what we need more than anything else at this point.

Re:Here's what Google needs

[node 3]

This:

Android eschews X

and this:

Third, targeting Microsoft must not be the aim, it must be the unplanned outcome. The aim must be tp "please" we the users.

Are the two most critical things that needs to happen for Linux to begin to take on significant market share. These are two of the biggest influences on the increasing success of Mac OS X.

Re:Here's what Google needs

[chabotc]

That's entirely missing the point. HTML5 gives you a very nice toolkit for building web apps allowing you full access to the computers computing resources with web workers (threads), storage and caching and graphics through canvas and even 3D graphics through O3D. The speed of the platform has also increased tremendously, in a year it's pretty much tripped thanks to FF3.5, Safari 4, Opera and Chrome. (and other goodies like location and no-plugin-required video playing)

The end result is that a web app can now approach a desktop app in features and speed, and with that you can finally stop worrying about what OS people run, that's becoming irrelevant, as long as they have a modern browser that supports HTML5, they can run your app. It also means that if you have a great idea, you can code it up and deploy it to everyone with a modern browser without having to ship a single CD or making a user go through a installation process

Forget about the OS, it's all about the apps! :)

Re:Here's what Google needs

[dustice]

I feel compelled to mention that your third criteria for success is also one of the core tenets of Google's philosophy - "Focus on the user and all else will follow." http://www.google.com/corporate/tenthings.html

Re:Here's what Google needs

[mcvos]

Forcing a feature into a standard to support your own proprietary application is shoddy.

Google Wave is open source, not proprietary.

Re:Here's what Google needs

[node 3]

What do you even care if Linux takes on significant market share? As long as your preferred operating system remains available to you, what do you care which operating system other people prefer?

Community is important. As they say, "no man is an island".

Re:Here's what Google needs

[badkarmadayaccount]

..Exchange...

I think I just figured out where Google Wave is going...
Any volunteers for a free/open Outlook connector for that?

Re:I'll give this much to Google

[Anonymous Coward]

Especially with all those *private* API calls only available to Micro$oft developers, leaving all other 3rd party Windows developers in the cold. Oh - and add to that, changing the APIs at every version / update, with no warning... gotta love that...

HTML 5 + Gears + GWT: resounding maybe

[Anonymous Coward]

I've seen the HTML 5 in-browser/no-plugin demo at Google I/O and of course it look impressive.
Now Google has to make a convincing case that the HTML 5 / Gears / GWT combo opens the world to developers who target the desktop. So far, GWT as great as it is, hasn't exactly gathered a big followship. At any rate, developing for the DOM presents some challenges and pitfalls, and IE 6 remains a major target that needs to be covered. That many corporations and casual users haven't "upgraded" yet, now plays out as an advantage for MS.
Even harder will be the mobile environment, where connectivity and bandwidth are weak and mobile web apps will require considerable optimization to not end up useless bandwidth hogs.

Re:HTML 5 + Gears + GWT: resounding maybe

[node 3]

IE 6 remains a major target that needs to be covered

No, it doesn't. IE 6 is a ghetto, and can be safely ignored. Anyone who currently uses IE 6 and either will not or can not upgrade to a modern browser is someone who isn't terribly concerned about using the types of apps that things like HTML 5 and Gears are meant to make possible.

Re:HTML 5 + Gears + GWT: resounding maybe

[mgblst]

The problem is there are large companies locked into that piece of ****. It will cost them loads to move on from there.

Re:HTML 5 + Gears + GWT: resounding maybe

[Anonymous Coward]

That's an unfortunate side effect of having invested in what is essentially proprietary and (now) 5 year old software. Non-standard software and protocols become obsolete if they're not supported; and IE6 isn't even supported by Microsoft anymore. I agree with what you've said, they bought the vendor lock-in and now the vendor screwed them. Tough. Someone needs to take IE6 out back and put a mercy bullet in it, it's dragged on for far too long.

Re:HTML 5 + Gears + GWT: resounding maybe

[Nekomusume]

Make the aps good enough, and don't even try to make them IE6 compatable. Just say "you need to upgrade for this to work. here's how you do it."

The main reason all those people haven't upgraded, is that just about everything on the web still works in IE6.

IE6 compatability is a bug, not a feature.

Re:HTML 5 + Gears + GWT: resounding maybe

[acb]

The corporate world is 90%+ wedded to IE, and much of it still stuck on IE6. (And it gets worse; one project I worked on last year had to work with an enterprise-wide "secure" deployment of an ancient, extra-buggy version of IE6 not seen outside of that company; we ended up losing much of the AJAX bells and whistles.) I'm told that this is because one thing Microsoft do better is remote administration of large enterprise-wide setups.

Which means that, unless there is a paradigm shift in the corporate culture, HTML5 will remain irrelevant on the corporate desktop; MS don't want to support HTML5 for the reasons described in the article, so IE support for these standards will remain either nonexistent or patched on with layers of hacks (like Google's JavaScript SVG library for IE8). Google's best bet is to ensure that such a shift happens.

Re:HTML 5 + Gears + GWT: resounding maybe

[exKingZog]

Very true. Local councils, especially, seem to only run IE6. There's a persistent 15-20% of our users who remain stubbornly on IE6 - this hasn't changed since this time last year, and all the inroads being made by IE8 are at the expense of IE7.

Re:HTML 5 + Gears + GWT: resounding maybe

[Just Some Guy]

Anyone who currently uses IE 6 and either will not or can not upgrade to a modern browser is someone who isn't terribly concerned about using the types of apps that things like HTML 5 and Gears are meant to make possible.

The first time Youtube's legacy Flash interface has significant downtime while the HTML5 version is still online is the day that the CEO orders an upgrade past IE6.

Re:HTML 5 + Gears + GWT: resounding maybe

[david_thornley]

If I had mod points right now, you'd get +1, Insightful.

For a business, the question is not the market share of a target platform, but the market share among those likely to spend money. There may be plenty of people out there with six-year-old computers, but they're likely not the sort of people who will buy expensive new hardware or software.

A person, or business, still on IE 6 is likely not a promising customer in any event. Depending on what you're selling, it's likely better to appeal to the part of the marketplace that's more likely to buy something you produce.

Re:HTML 5 + Gears + GWT: resounding maybe

[vitaflo]

The problem is there are large companies locked into that piece of ****. It will cost them loads to move on from there.

Evolve or die.

Re:HTML 5 + Gears + GWT: resounding maybe

[node 3]

The corporate world is 90%+ wedded to IE, and much of it still stuck on IE6.

That's their problem. As the web moves forward, they can stay in their IE 6 + Win32 ghetto if that's what they want. It's foolish to think the world should hold back on progress simply to cater to these idiots.

Hate to be a spoilsport but

[gcnaddict]

Microsoft actually contributed lots to HTML 5, at least according to Chris Wilson (Software Architect for IE)

In effect, it's like semi-Microsoft v. completely-Microsoft. (food for thought)

Re:Hate to be a spoilsport but

[pizzach]

You better be careful. The people who think that W3C has no relationship with MS/IE and that it only does things to hurt IE will get mad. :-p

Re:Hate to be a spoilsport but

[drinkypoo]

Microsoft actually contributed lots to HTML 5, at least according to Chris Wilson (Software Architect for IE)

Someone had to introduce problems, incompatibilities, and inconsistency or it wouldn't be a proper standard.

Re:Hate to be a spoilsport but

[gcnaddict]

Who said he was the only one to contribute? I said Microsoft contributed, not Chris Wilson.

Re:Hate to be a spoilsport but

[node 3]

I'm not sure what you're getting at. If the idea was W3C vs MS, you'd have a point, but this is Google vs MS. The fact that Google is using a tool that is being developed with help from MS is somewhat ironic, but doesn't make this a "semi-Microsoft v. completely-Microsoft" battle any more than the Japanese attacking China with gunpowder weapons is "semi-China v. completely-China".

Re:Hate to be a spoilsport but

[chabotc]

If only IE would support silly little things like the canvas and video tags, or have proper SVG support for that matter.

They have stated they intent to support HTML5, but I'm still waiting for this to actually take shape (hope they will!)

Re:Hate to be a spoilsport but

[mcvos]

Wow reading that sure ticked me off. Microsoft basically just said that they suck and cannot ever implement any spec correctly, therefore they need version tags in html.

From what I understand, they're just claiming HTML 5 is not the same as HTML 4 or HTML 3, and the document should identify which it's using. That sounds perfectly reasonable to me, and it has always been the standard so far. Doing away with it would be stupid.

Never thought I'd see the day where Microsoft is defending the importance of proper DOCTYPEs and the open source community is ignoring it.

Re:Hate to be a spoilsport but

[mounthood]

Microsoft actually contributed lots to HTML 5, at least according to Chris Wilson (Software Architect for IE)

Someone had to introduce problems, incompatibilities, and inconsistency or it wouldn't be a proper standard.

When you say "HAD TO" what exactly do you mean?

Re:I'll give this much to Google

[node 3]

Microsoft is where it is today because it is the easiest OS for third parties to work with.

That's not even remotely true. I'm curious as to how you came to this conclusion? Are you simply comparing Windows with Linux (and most any other X11 based system)?

The reason for MS's success (specifically, with Windows) is due to developers targeting the dominant system, and Windows became the dominant system primarily through being installed on the overwhelming majority of PCs. None of this was based on being the most "developer-friendly".

Who still listens to these two clowns?

[Anonymous Coward]

Two guys gave themselves a fancy name ("Open Document Foundation"), which opened them doors to some panel discussions. They don't have any role in ODF standardization. All they did in the last couple of years was to act like little MS shills. Last time they attacked the Open Document Standard their role was quickly uncovered.

Their insignificant role does not deserve any attention, but as they trolled their way into mainstream media IBM's chief ODF architect Rob Weir was bothered enough to discuss the technical merits of their "contribution" .

More Automated Spam?

[drinkypoo]

This comment is way too similar in style to this other comment. I call shenanigans.

Re:More Automated Spam?

[Max Littlemore]

Yep. I reckon we need a "-5 SPAM moderation".

Re:I'll give this much to Google

[mR.bRiGhTsId3]

As a "hobbiest" developer working on a cs degree, my personal experience is that windows is infinitely easier to develop for. It removes the need for me to understand 3 or 4 different arcane scripting languages that get together and have a preprocessor orgy with themselves in order to output makefiles. This has been my experience with autotools, scons, and cmake (cmake wasn't too bad on the whole). On the other hand, I was able to get a standard copy of VS 2008 from my university's campus connections and I can configure building things much easier. That and WPF is a nifty gui toolkit. I found myself able to hand write the gui description files after about 1/2 an hour of tinkering. Something I never could have done with glade or qt's ui files.
For those of us who like to tinker on little projcts on our own, if we can hardly set up a build system, and the docmentation for said build system is horrible, whats the point (I'm looking at you autobook).
That being said, I am curious what you do consider to be the most developer friendly system, particularly if you have experience in industry.

Gary Edwards?

[bmo]

Hey, uh, wasn't he one of the ones that threw a tantrum (along with sam and marbux) when he didn't get his way with preserving Microsoft "dark matter" (undocumented RTF encoding) in ODF and then proclaimed that ODF is doomed to fail and all that nonsense when everyone told him to stuff it where it doesn't shine??

I am shocked. Simply shocked to see that he's extolling Microsoft's "virtues".

Nothing to see here, folks, just another softie trying to sabotage open standards by throwing chairs at it.

--
BMO

Re:Gary Edwards?

[Just Some Guy]

Hey, uh, wasn't he one of the ones that threw a tantrum (along with sam and marbux) when he didn't get his way with preserving Microsoft "dark matter" (undocumented RTF encoding) in ODF and then proclaimed that ODF is doomed to fail and all that nonsense when everyone told him to stuff it where it doesn't shine??

Yep. And also one of the ones who created the OpenDocument Foundation and then decided ODF (the format) sucked and that ODF (the organization) wanted to support something else.

Re:I'll give this much to Google

[node 3]

I pretty much pointed out that Linux/Unix/X11 aren't necessarily the best example when I wrote, "Are you simply comparing Windows with Linux (and most any other X11 based system)?"

That being said, I am curious what you do consider to be the most developer friendly system, particularly if you have experience in industry.

Presently, OS X is extremely easy to develop for. In the past (the context here, after all, is MS's success, so you have to look at what came before), both OS/2 and BeOS were supposed to have been fairly advanced from a developer point of view, as was Nextstep.

Even further back, comparing Macintosh System, AmigaOS, etc., with DOS and somewhat later, Windows, is relevant. I really don't think ease-of-development played a significant factor so long as development was "easy enough". Commercial interests are a much greater factor.

OS-less netbook

[OutputLogic]

Hypothetically speaking, if there is a powerful Java processor that runs Java Virtual Machine (JVM) in hardware, and a browser application written in Java, you'd get an OS-less netbook

- OutputLogic

Re:OS-less netbook

[shutdown -p now]

Java standard library deals with concepts such as threads, files, network sockets, graphic devices etc. In an "OS-less netbook", how are those going to be implemented - magic fairy dust?

Desktop?

[gmuslera]

Google dont need to displace Microsoft from the desktop. What google is doing is displacing the desktop itself. Once you have the same info and roughly the same functionality from your cellphone, netbook, computer, gaming device, whoever else computer and so on, "Desktop" is becoming meaningless. Microsoft must give away the desktop and embrace the cloud to have any chance, just because it isnt a battlefield anymore.

Where is the foundation on which you build?

[westlake]

Chrome. Safari. Firefox.

"The edge of the web."

God alone knows what that means. Market share dominated by the home user and the enthusiast. Chrome very immature.

Internet Explorer. The browser you use at work. Rich tools for deployment and management by the system administrator...

In the simplest terms:

You can build a business ground up from the loading dock and point of sale to the clerks in accounting to the guys and gals in middle management and the executive suite and never leave the working environment of "MS Office."

That has enormous implications for recruitment, staffing and training at every level.

Cloud vs Desktop? Aren't they the same?

[caywen]

Isn't the deskop really just the next evolution of the cloud? Once the desktop becomes an active participant in the cloud? I think the next step will simply be to make all your desktop apps available anywhere. We're just about there already with remote desktop connections. Isn't the path of remote desktops and virtualization just as valid a distributed computing model? In the future, there might be so much bandwidth and parallel computing power available, a single server could serve remote connections to thousands of simultaneous virtual Win7/OSX/Linux machines. And you won't have to actually rewrite OpenOffice 10.0 for web.

Collaboration

[drx]

If the cloud would only be about data storage there would be no advantage over a Desktop app that saves to my hard drive.

However, Desktop Software is totally behind when it comes to collaboration. I have sent enough "DOCs" around and received them back and edited them again and sent them around again to understand that it sucks badly. I have enough of "can you send me the latest version of ..." and welcome online apps to solve this gigantic and ridiculous problem. Of course i would prefer to have Desktop apps that do the same thing, but as it seems at the moment nobody can get their act together and do real time collaborative Editing in a way that is more meaningful than Gobby. :)

Re:Collaboration

[jonaskoelker]

So, where's the online MagicSetEditor? Or the on-line sound editor? Or the on-line wesnoth map editor (FOSS turn-based strategy game)?

Depending on the data format, you might be able to collaborate with client-side apps and version control with on-line storage. If you can convince your friends or co-workers to use version control.

If what you care about is collaboration, all you need the cloud for is a place to store the data that everyone agrees on. And that's only if you can't have your cloud be in-house.

You don't need to edit your data in your browser. You don't even need a browser to store your data on-line.

(for binary/non-diffable/non-mergeable data, you might need exclusive per-file locking...)

Re:Collaboration

[drx]

Depending on the data format, you might be able to collaborate with client-side apps and version control with on-line storage. If you can convince your friends or co-workers to use version control.

I am sure Google Docs uses diff as well, and Ghostscript and everything. The good thing about them is they made it into a software i do not have to convince anybody of, it is usable as-is.

If there would be a better alternative, not relying on Google's merits, i would use it immediatly.

You don't need to edit your data in your browser. You don't even need a browser to store your data on-line.

(for binary/non-diffable/non-mergeable data, you might need exclusive per-file locking...)

I know *I* don't need anything fancy for editing stuff, however, everybody else i am collaborating with has seems to have needs different from mine.

AND NEVER MENTION EXCLUSIVE FILE LOCKING AS A SOLUTION AGAIN!!!1 ;)

Amazingly we should side with... Microsoft!

[Nicopa]

The standard desktop is better than Google desktop. Yes, everybody says, to put Google in a good light: "standard compliant" browsers, but that means nonstandard compliant mail, nonstandard everything else. We won't own software, we'll be always customers, dumb terminals, served from huge company's "clouds". Free software will be over, irrelevant. We won't be able to improve and modify our environment, we can't improve Gmail ourselves, there's no alternative/better/innnovative client for Gmail.

Economic forces are taking technology down a terrible path. The past is better: a world of protocols, servers and clients. A common neutral space...

The "portable" desktop, having your data everywhere should be solved by other means... I don't know, perhaps we should have personal servers, or at least we should contract personal servers from some kind of "personal server providers", which should be a standard and non-monopolistic thing. The "presence providers" envisioned by the XMPP protocol comes to mind...

Re:Amazingly we should side with... Microsoft!

[Lonewolf666]

The standard desktop is better than Google desktop. Yes, everybody says, to put Google in a good light: "standard compliant" browsers, but that means nonstandard compliant mail, nonstandard everything else. We won't own software, we'll be always customers, dumb terminals, served from huge company's "clouds". Free software will be over, irrelevant. We won't be able to improve and modify our environment, we can't improve Gmail ourselves, there's no alternative/better/innnovative client for Gmail.

(emphasis mine)

The points in boldface are a reason to side with the Open Source community. Neither Google nor Microsoft is any help on these aspects. And IMHO Ubuntu and similar systems are already quite usable for most purposes.

Re:Amazingly we should side with... Microsoft!

[Jane_Dozey]

Perhaps this is where social networking sites will evolve to: service providers for our information. It wouldn't be that greater step for the facebooks and myspaces of the world to add a bunch of servers and rent space on them with interfaces that can interact with cloud services. After all, it seems that most people already have an account and use it in order to store various photos and information about themselves.

If the technology for interacting with the cloud is open and well documented it'd make sure those who don't mind leaving their documents, data and information in the hands of 3rd parties can do so, while leaving the option to do it yourself open.

I'm quite cynical about the cloud at this point as there isn't nearly enough of a global network to make me want to have something relying on a connection to work. However, I think it's a nice idea for the future, if implemented sensibly and away from any political or corporate interests.

Re:Amazingly we should side with... Microsoft!

[AnyoneEB]

Agreed. Ironically, Google Wave might be [part of] the solution.

Having your data be portable really should not be a problem. Storing it (encrypted?) in a Wave would be one way. Really it seems like you should be able to have your data automatically get replicated (encrypted) across all of your friends' computers (somehow registered with your own) with the assumption that it is very unlikely all of them would be down when you attempt to retrieve a document. Of course, that would require that all of your friends' computers weren't NATed.

To be fair, you can "improve" your own GMail experience via GreaseMonkey scripts. There are a few Firefox extensions which are basically bundles of scripts to add various features/tweaks to GMail. On the other hand, GMail is still very much so non-free with the very major issue that you cannot use it without letting Google see all of your e-mail. As far as I know, SquirrelMail is its closest competitor.

Re:Amazingly we should side with... Microsoft!

[MikeBabcock]

You're actually seeing this the opposite way that I do. If most of the software my users want to use is served by the cloud, then I can install Ubuntu on their desktops without fear of compatibility issues with the software they want to use. Windows becomes less relevant as fewer people use the Windows API to write their software, and this strikes fear into Microsoft.

It used to be people asked me if their computer could be used to run Microsoft Office, and now they ask if it will load Facebook and play their favourite flash and Java games.

Re:No, they want to make something /good/

[Lord Bitman]

OOo is not a good starting place. They already have chrome + gears, which is more than enough to use google docs by itself.

Re:I'll give this much to Google

[X0563511]

You realize that preprocessor "orgy" happens anyways, just behind the scenes? And that if you used a proper IDE, you get the same "blinders" that MS-VS gives you?

And if that is your only complaint - the tools, then I challenge how you can call yourself a developer. Have you SEEN the Windows API? Now... thinking of that, have you seen glibc?

Please forgive me if I'm wrong, but from what you've given us, I'm hesitant to take anything you've said seriously.

The cloud is here

[mcrbids]

OOo has been able to 'save to the cloud' for a very, very long time. WevDAV was introduced many years ago, and works as well today as ever. It gives decent security, is quite reliable, and can be seen as a local drive on most modern OSs. (Sadly, even Windows Vista still needs NetDrive)

Bottom line: there is no need to NOT save to the cloud in basically any program out there today, client-based or no, if you are at least somewhat intelligent about it.

Re:HMTL?

[hvm2hvm]

Well he was reading the second part as "inclod sensitive you" and he probably didn't know what it meant.

Re:I'll give this much to Google

[moon3]

Exactly, look at Google, they offer you "Eclipse" and "Java" for production. Not even native code support.

Suddenly Microsoft looks super friendly.

Re:HMTL?

[MrMr]

I have to ask: Which words did you fail to miss-spell?

Astroturf from Gary Edwards

[Anonymous Coward]

Oh come on, Timmothy. Edwards has already been discredited for his astroturfing fake ODF news which he ran under the "Foundation" he's now moved his shit to Facebook and others. His foundation actively lobbied against ODF. He was a shill then, or at least one of Bill's "Useful Idiots", and he's the same now.

Re:I'll give this much to Google

[alien_life_form]

You realize that preprocessor "orgy" happens anyways, just behind the scenes?

Yup, it does. So does conversion to machine code and conversion of logical information to electrical signals. What's your point?

And that if you used a proper IDE, you get the same "blinders" that MS-VS gives you?

Taken at face value, the OP is complaining of poor IDEs on LAMP platforms, so I would say he understands this, also.

And if that is your only complaint - the tools, then I challenge how you can call yourself a developer.

As a complaint, it is not trivial (it has also been beaten to death over the years, but that is a different matter). Developers spend most of their time inside tools. Masons are less effective if deprived of a good hammer. Having used both Eclipse and VS, I have (and hate) to admit the the latter is (was: it's been a couple of years) still a frontrunner. Intellisense, just to name a feature, is a huge time saver for anybody not knowing the docs by heart.

Yes, you can 'vi foobar.c'. Even cat > foobar.c is known to work. As is do-it-yourself eye surgery using an old pen and a rusty can opener. Most of these practices are not advisable, though.

Have you SEEN the Windows API?

I don't know about you, but, to SEE APIs, I require some amount of controlled substances, heavy drinking, or possibly both.

Cheers,
alf

Does HTML 5 away with meaningful DOCTYPEs?

[mcvos]

Chris Wilson joined the HTML 5 working group (WHATWG) in April '07. Over one year, his sole contribution was that HTML versioning crap: ref.

I just read that and the WG discussion it links to, and from what I understand, Chris Wilson wants to maintain html version numbers in the doctype declaration, while others want to pretend html 5 is the only version of html in existence, which is completely stupid. I hate MS as much as the next guy, but Chris Wilson is completely correct here.

I haven't really been following the HTML 5 standard much so far, but I guess this means that HTML 5 has already done away with meaningful doctype declarations (which also helps to explain why a w3c acquaintance of mine doesn't like html 5 much).

Mod those hairy feet up!!

[Runaway1956]

I seldom look to see if I have mod points - for this post, I looked. Saving to the cloud is all well and good, for convenience, I guess. But NOTHING beats the combined reliability and security of my own backups. When convenience takes a shit, that old reliable external drive is still sitting in the corner, or in the back seat, or under the airplane seat, whirring away, and waiting to go.

Yeah, I do save some things to Gspace. But I certainly wouldn't save anything like a confidential industrial secret document. Encrypted on my hard drive is the ONLY place to put it!!

Correction...

[TheLink]

Oh... looks like the 5 minutes was to download a new version of Java or something.

And IE doesn't do that prompting, whereas Firefox does (at least on my setup).

Re:Correction...

[bami]

The thing I hate about webstart (happened to me at least), is that it really "installs" software, it adds registery keys and adds a program to the software panel in the configuration panel in Windows.
I'm all for java or webapplications, but that is where it crossed the line for me. Not to mention the prompt in firefox.

Re:Correction...

[dna_(c)(tm)(r)]

[...]but I doubt "corporate" java apps are going to be small[...]

I would think that in a corporate environment, that would not be such a big issue.

Now, I'm not advocating to use ws/jnlp for everything, but for complicated graphical tools, why not? It's an alternative.

Oh... looks like the 5 minutes was to download a new version of Java or something.

I think that's a good point, one more deployment issue out of the way...

Re:Abstraction

[Keeper Of Keys]

It's a shame you had to mar the good points you made with an uninformed comment about CSS and Javascript. Done properly, they are both powerful and, yes, elegant tools.

Gobboldygook

[omb]

This comment says almost nothing:

The Office game is almost over, OOO is good enough, and existing anti-trust, most recently in Russia increases the challenge.

AD and Exchange are currently coporate lock-ins and the Open Source community must look for its commercial partners, Google, IBM and Oracle to help fund drop-in replacements.

It is very unlikely that MS will get a lot of traction with a new proprietary program, especially outside the USA

or browser reliabilitiy without crashing, cough FF

[cheekyboy]

Pick a tab any tab, watch it suck all cpu because stupid advertising companies write bad JS that while loops forever killing all other tabs.

Re:I'll give this much to Google

[Late Adopter]

I don't wish to contradict you, since you raise some good points about Linux (autotools... *shudders*)

That said, if you *are* interested in developing for Linux again in the future, give Qt a shot. QMake is by far easier than any of the things you've listed, and Qt Creator is a great IDE that uses qmake files as its native project files.

Come to think of it, you could do this under Windows also.

In the best of all worlds...

[davide marney]

I'd have a permanent historical archive of everything I ever created or edited. My archive would be physically accessible to me and my designates alone.

On an automated schedule, a complete backup copy of my archive would be stored on another physical device, again accessible only to me.

Some subset of my archive would be synchronized to a cloud service, so I could get to my data from anywhere. Some smaller subset of that would be publicly available so I could share.

Getting data into my archival stream would rarely require explicit action on my part. The software I'm using would be aware of the existence of my archives, and would automatically forward version updates down the chain. No matter where the edit starts, eventually the data winds up in the private archive that I, alone, control.

Naturally, my historical archive includes every email I've ever written or replied to; every photo I've ever taken; every document or spreadsheet or memo; every comment on Slashdot; every journal entry; in short, my entire, digital record.

Re:or browser reliabilitiy without crashing, cough

[maxume]

It could be a platform thing (I'm running FF3 on Windows), but you might want to try disabling flash (or running flashblock); I don't ever notice javascript based ads locking things up, but something like 2/3 of all flash objects that I activate eat 100% cpu.

Re:I'll give this much to Google

[mR.bRiGhTsId3]

No, I'm pretty sure that msbuild works absolutely nothing like autotools does. Autotools requires the following steps if memory serves:

1. Write config.ac in its specific scripting language using 2 or more sets of expansion macros the AC, AM, and AUTOHEADER stuff
2. Write Makefile.am using a different syntax that sort of looks like a normal makefile but not really.
3. run autoconf
4. run automake
5. run configure

And the so called "blinders" that you can get out of the IDE's don't seem to actually work. My favorite was that for a really long time after KDE was coming out and the recommended build system was cmake, KDevelop didn't actually support cmake. My other experience with an old version of Anjuta could be summed up with, "Here! We're an awesome IDE with support for autotools. Instead of typing arcane expansion macros directly into a text file, type them into this oddly named dialog box instead." In all fairness, I haven't tried anjuta in a while.
On the other hand, msbuild takes a single xml file and runs the appropriate tools. I don't have to remember the obscure syntax to add header defines if I don't want to, I just have to add -D, -I, or -l options in a box. Granted the whole linking to .lib instead of the actually .dll is a little wierd
Each system has its own set of hoops to jump through. I posit that you think glibc is so much better because it is what you are used to. Keeping that in mind, lets think about it. Which function do you think is better self documenting. CreateProcess or fork. Or the other things that spring to mind like symlink vs CreateSymbolicLink (yes newer NTs supports symlinks).

Google the evil one now?

[cjjjer]

1. Make a desktop that uses a browser for it's application container
2. Make a browser with a fast JavaScript engine and HTML5 to make a better UI
3. Make a framework to support applications made for said browser
4. User experience is now dependant on said apps and browser
5. ???
6. Profit

Seems pretty clear to me what Google is trying to do, something MS has already tried and almost succeeded in. Except Google does not want a market share to gain profit they want advertising dollars for profit. Is there any difference?

Microsoft fights back

[Akir]

As it turns out, the early builds of IE9 are starting to implement HTML 5. For instance, they support the video tag. However, they made a statement that they are only planning on implementing the intel INDEO video codec, with PWM audio. And they're finally supporting CSS and SVG propperly, but only if the color scheme matches the scheme they chose for Windows 7. And after years of debate, they now support an early draft of VRML.





I'm obviously lying, since Microsoft always plans on the "let's not improve the product until it's at least 10 years behind the times" plan.

Sadly, though, The way I described their new version of IE is waaaaaaaaaaaay too realistic.

Re:I'll give this much to Google

[Tangent128]

For command line applications, sure, POSIX is simple.

But last I checked, glibc doesn't offer much in the way of a GUI.

Re:I'll give this much to Google

[SlappyBastard]

"The reason for MS's success (specifically, with Windows) is due to developers targeting the dominant system"

And how did MS become the dominant system?

Sometime way back in the foggy mists of history, a bunch of people made a bunch of operating systems. Someone would say, "Wow, Mac OS is awesome!" And if you ever tried to write something for Mac System 7, you ended up praying God would kill you soon.

For some incredibly strange reason, a whole bunch of programmers said, "Screw that!"

And then they did the same thing with a Windows-based system. Only, this time they could make it work without pulling their eyeballs out.

We got to this point because MS was the company whose OS was easiest to work with.

It's cute to say (in a later comment) "presently OS X is wicked super awesome jesus wowswers to work with". But, that is to be oblivious to how we got to this point. For waaaay too many years, Apple's OSes were criminally evil from a developer-friendliness standpoint.

Thanks for troll modding

[SlappyBastard]

Does anyone on Slashdot ever feel just the slightest bit retarded for being reflexively anti-Microsoft?

Re:Thanks for troll modding

[HW_Hack]

No I don't think so ... a lot of slash-dot readers have been around a while and seen M$ repeatedly deliver totally dicked-up solutions - Or something like Vista which is total marketing candy and a technical clap-trap.

When M$ can go 5-8yrs delivering real robust and innovative solutions then thats fine. Frankly I don't see that happening soon

Re:I'll give this much to Google

[VGPowerlord]

Even back in the mid 90s, it was more common for Windows developers to use MFC rather than access the API directly.

Re:I'll give this much to Google

[Super_Z]

The autotools package is designed to make source code easily portable to different UNIX systems. Visual Studio is an IDE for creating a program on the Windows operating system. You (and the GP) are comparing apples to oranges.

Re:I'll give this much to Google

[node 3]

There were other OS's than Macintosh System, many of which ran on the same hardware as Windows. Some of the contemporaries of Windows, many predating Windows (especially if you start with Windows at 3.1, which is the first version to have any real success).

Commodore mismanaged themselves, OS/2 was mis-marketed by IBM and sabotaged by MS. DR-DOS was sabotaged by MS. GEM and DESQview were simply outclassed by Windows. X11 was rarely promoted outside of Unix. BeOS was too set on replacing Mac OS. And I haven't even covered half of the OS's out there.

Commercial forces far outweighed ease-of-development, which is merely one of the many commercial forces, and not even remotely the most significant one.

Re:I'll give this much to Google

[shutdown -p now]

And if that is your only complaint - the tools, then I challenge how you can call yourself a developer. Have you SEEN the Windows API?

You don't need to use Windows API for many kinds of applications - .NET does the job just fine, and has reasonable object-oriented APIs for most everything.

As for tools - good development tools (IDE, debugger, profiler) can reduce development time manyfold. Apple and Trolltech also understand it. All Java guys understand it (in fact, Java still probably has the best tooling overall, to date).

Re:I'll give this much to Google

[alien_life_form]

"The reason for MS's success (specifically, with Windows) is due to developers targeting the dominant system"

And how did MS become the dominant system?

Two words: cheaper boxes.

Back in the days, Macs commanded and integer multiple of the price of an x86. Unixes, think tenfold and more. This does not even begin to address the idiotic API wars Unixes were waging: System V vs. BSD idioms, vastly incompatible (not to mention ugly) GUI toolkits.

But all that is irrelevant compared to the price tag issue.

Cheeers,
alf

Re:I'll give this much to Google

[cheekyboy]

nothing wrong with the windows api, at least you know its there, unlike 90% of linux apis which might not be there depending on the distro or version of the library which changes to often, or new releases break old things.

One thing MS does right, is keep ALL old apis, so nothing old fails, linux or OSS guys usually dont care, and say tuff luck, N api is going away, live with it, they have less understanding of their customers, and simply think that 1000000s of people will 'adjust' their code to the new ways. Well if you coded the original properly you wouldnt need to redesign stuff would you.

the missing TCO

[cheekyboy]

I guess that is one TCO component MS never mentions, the cost of being locked in, then moving to the next major version that breaks all your existing badly developed systems.

Re:I'll give this much to Google

[MikeBabcock]

Microsoft was very wise to put a lot of money into its development platform. It spent a lot of money building up Visual Studio and other developer supports like the MSDN. As a result, a lot of third parties wrote software for the Windows platform because it had a low barrier to entry and was easy to work with initially.

Of course, many developers eventually hit the wall where they run into limitations of the API they've chosen as I did working with Windows, and some choose to spend many hours working around it instead of jumping ship. In the end, there's a lot of software for Windows by a lot of amateurs (the huge shareware libraries) and eventually quite a bit of professional software too.

Re:I'll give this much to Google

[MikeBabcock]

Oh man, I just had this horrible taste in my mouth. It must be the networking applications I was writing with the MFC that came back to haunt me. Man that was horrible.

Re:I'll give this much to Google

[MikeBabcock]

Nor should it.

Does the Windows "C" library offer a GUI? The GUI is provided the same way as on Linux and other platforms, but more integrated into the API most programmers use (which isn't the C library).

MFC, .NET, etc. are all abstractions with their own libraries. You can just as easily write code in C++/Qt for Windows and it won't look like a typical Windows app at all.

Re:I'll give this much to Google

[MikeBabcock]

By that token, write it in Java and you get an IDE and no need for the Windows API either.

Your comment isn't really a selling point of .NET specifically.

Re:I'll give this much to Google

[shutdown -p now]

Yes, sure, and Java should be considered, alongside Qt, wxWidgets, and other cross-platform toolkits, when speaking of Win32. Point being that you get the same stuff as on Unix, but the tools are still better.

Note that original topic of the thread wasn't Win32 vs Unix APIs, or VS vs Emacs, or whatever. It was whether development in Windows - in general - is friendlier than development on other platform.