• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Cloud computing- Your opinion.

IBM had their time-sharing concept in the 70s. That allowed users access to their mainframe and shared resources and applications for a fee, metered out like a utility.

Software-as-a-service is becoming economical due to bandwidth, and it works well for some applications and needs.

The company I work for bought a ton of these little guys:
VSERIES_WKBD_MSE_LG.GIF


This is a ThinClient. It cost about $150 for one. If it breaks, you swap it out in a few minutes. It's got a firmware chip on it, and everything else is piped in through Ethernet or WiFi. They aren't great, but they're cheap. Thin clients are alive and well if you have the bandwith for it.
 
Well, I happen to think Windows Azure is an excellent name for Microsoft's cloud computing platform. It's name almost sounds like the word "assure", while simultaneously sounding like a nice color scheme for one's desktop.

And we all know that the better-sounding names of operating systems tend to be Microsoft's best. I mean... look at Windows Vista! Right?!

This might qualify as sarcasm.
 
Heh - everyone has been hopping on the Cloud name bandwagon.
Nimbus, Cloudburst, SilverLining, Cumulus, etc...

But, wait until this is a boring consumer offering we're all complaining about. It'll be "the fog" or "smog", or "thick cloud cover limiting visibility" or "40 days of rain" or...
 
Now ... that server is likely a virtual server, one of several virtual servers on a box that scale automatically and may even be mirrored on other boxes.

Again, this concept is not new, only it's attainability by the general public.
 
Azure... sounds like sure... sounds like, "plays for sure"?
 
Actually, a thought. I haven't watched that 45-minute video because I'm lazy. I also have not looked up anything since starting this thread.

However, my concern has always been that - for example - Google goes bust in a big Enronian explosion. What happens to your data? You should be keeping backups, either with your own infrastructure or with a third-party cloud backup solution (which solves the problem and doesn't mean discounting cloud apps!) and more importantly what happens to your virtual office?
Say you have a 100-strong team across the world, all using google apps. Then google dies. How long will it take you to convert everyone to some other system and re-upload the documents you backed up? With local apps that's not an issue.
Solutions like google gears help, but how prevalent are they? I've had issues collaborating on google spreadsheets (one client likes to use it as a bug tracking database substitute) where I'll be happily editing away and then suddenly the page will reload a different edit, losing all my changes. This might be uncommon, but it's still teething days for collaborative document management on this scale.

Lastly, persistence of availability of information. Cor, that sounded almost like I knew some buzzwords. I mean, some businesses are constrained to keep copies of documents available for a certain length of time. This made the headlines over the last few years with certain departments saying they couldn't use MS Office because it was a proprietary format and they couldn't rely on having access to it in 10 years (say) whereas with an open format they could. Well, with a cloud-hosted office, where do you stand? If you keep a download open-format version in sync for backup purposes then I suppose you're ok, but that rather begs the question, doesn't it?
 
What happens if your local computer gets blown up? At best it would take a few days to get another computer and get everything loaded. That is if you keep real time backups. If not you lose all the data since your last backup. And hope that your backups are any good.
 
However, my concern has always been that - for example - Google goes bust in a big Enronian explosion. What happens to your data?
Hmmm...

www.google.com/apps How secure is your sensitive business data?
Why is Google Apps secure?

Google operates one of the largest networks of distributed datacenters in the world, and we go to great lengths to protect the data and intellectual property on these servers. These facilities are protected around the clock and we have a dedicated security operations team who focuses specifically on maintaining the security of our environment. The controls, processes and policies that protect these data have successfully completed a SAS 70 Type II audit. There are three main components to our security practices:

  • People – Google employs a full-time information security team including some of the world’s foremost experts in information, application, and network security.
Call me naive... but I have a hunch they might have at least considered the risk of a "big Enronian explosion"... and maybe... just maybe... have a plan to mitigate the risk
 
I think it is perfectly possible to make backups of your files on your own hard-drive.
And doing so could probably be automated.
 
Yes, cloud computing is just the thin client idea. The difference today is broadband - prior to that it was too slow and unstable. Now we're used to always on high speed connections that are beginning to be more reliable than our own PCs.

If Google did suddenly go Enron, the very fact they'd have millions of users wanting their data as fast as possible would make it an extremely attractive takeover target. Did anyone actually have their gas or electricity or whatever turned off when Enron collapsed?
 
This isn't a new idea, just an old idea that didn't work without internet being revived now that most folks are online. When computers didn't talk to each other, this was much more complex, and failed as a business model as soon as computers became cheap enough for businesses to buy their own.

Here's how it used to work, more or less.

My father was one of National Cash Register's programmers, back in the day. He was responsible for support of a set of programs, run at several data centers, that calculated hospital payroll. The time card data from the subscribing hospital's employees arrived in crates, and went to data entry, where a huge team of 'key punch girls' converted it to punch cards. The punch cards were fed into a reader, and converted into magnetic tape. The tape was processed through the mainframe, which ran an automatic card puncher to generate a crate of output cards. The crate of output, together with a punched paper tape loop that had the check number series data on it, would be sent to printing services, which would print, crate, and ship the checks.

One of the first places I ever worked still used a system like this, and I remember how inconvenient and irritating it was. First, because of the time it took to ship, process, and ship again, they were one pay period behind. That meant that new hires had to wait a month before seeing any income, which I remember as being rather harrowing. Second, we were at the mercy of truck freight. It wasn't uncommon to show up on payday friday and be told 'well, the checks aren't in yet. The truck got delayed. Come back monday.' It was a miracle they never had a riot. Third, any mistakes on your pay stub would take a month to correct. Oddly, they never made a mistake in my favor, but plenty of mistakes in theirs.

A
 
Call me naive... but I have a hunch they might have at least considered the risk of a "big Enronian explosion"... and maybe... just maybe... have a plan to mitigate the risk

Yes, I expect they have. But it's still trust in someone else - and I'm not just talking about me here, I'm trying to second-guess what the man in the street might feel too.

What happens if your local computer gets blown up? At best it would take a few days to get another computer and get everything loaded. That is if you keep real time backups. If not you lose all the data since your last backup. And hope that your backups are any good.

The odds on every one of your offices around the world getting blown up at the same time are...? The concern I'm expressing is about adding another single point of failure.

To be clear, I don't really have these paranoias, but I think they should be at least mentioned. So I can say "I told you so" if it all goes tits up.
 
To be clear, I don't really have these paranoias, but I think they should be at least mentioned. So I can say "I told you so" if it all goes tits up.
Not if... when

Remember... its a binary world:
  • There are systems that have gone tits up
  • And there are systems that will go tits up
However... this ain't a reason to be afraid of cloud computing
 
Google apps offers a whole package of interesting "cloud" applications.

Theoretically you could run your whole office on Google apps, providing you had a broadband internet connection.
That's what we do here at work, actually. We use Gmail for email and use Google docs to share documents.

This is the main reason I'll stick to MS Office or whatever else will come by for the foreseeable future. It's not just the Gmail, it is also my ISP, my modem and my network card, all of which can fail. If any of that happens on, say, the evening of a long weekend (for example, thursday, dec 24th) and I need to have them done by monday, it creates an essentially unsolvable problem.

I know the solution isn't perfect, but it's better to rely on as few failable devices as possible, IMHO :)
Your hard drive can fail. Even if the chances are low, they're probably higher than Google and your ISP (unless your ISP really sucks) remaining unavailable for a long period of time.
Also, nothing stops you from downloading backups from Google Docs if you need to. Plus, Google docs have version control too, which is reaallly nice if you accidentally overwrite data.


(No, I don't work for Google... though at times I kinda wish I did :P)
 
I'm guessing you can buy a license if you want to run it on your intranet. Running your own cloud makes sense, and one more nail in Microsofts coffin. (not dead yet, though)

Hehe - not even close. They made $13B in revenue last quarter and only $2.6B was from consumer OS stuff. They realized long ago - like most SW companies - that the real growth potential was in middleware and B2B, not in consumer end-user applications. In fact, that's the only MS division that grew YtY in their last quarterly report.

But, since the initial impact of cloud computing is in serving up end user apps you can expect MS to be a big player. Their Word, Powerpoint, Excel, etc... are the international standard (I work for a competitor, and internally we use the MS products rather than our own). Love them or hate them, everyone knows how to use them by default. And with their OS on so many PCs they can easily establish a SaaS cloud directly accessed from the OS - from which they can serve up all their end user apps for a small pay-as-you-go fee.

Cloud computing - in its initial phases as a way to deliver consumer Apps - will line Microsoft's coffers bigtime.
 

Back
Top Bottom