Can’t add more, though, I agree with what you’re saying!

For the CMS example, yes, when looking at the actual data itself it is the responsibility of those in charge with the data store management.

I think it comes down to accessibility vs. availability. Most technology companies wave the big availability flag which although crucial, is not the important one. From a user point of view, accessibility is the only thing that matters. If the data is available (as in the linkedin case), yet completely unacessible…

We’ll explain it this way:

Avail.(1) + Access.(1) = Happy Customer (2)
Avail.(0) + Access.(1) = Upset Customer (1)
Avail.(1) + Access.(0) = Frustrated Customer (1)
Avail.(0) + Access.(0) = No Customer (0)

I think the reality is that in order for services like LinkedIn to work, we have to trust both the service accessibility, as well as the data availability. Sure, large companies put quite a bit of funding into data storage, yet it is still part of that company. If the data is lost (for whatever reason), what recourse do you have? You used their system, you stored it there, yeah they lost it, but didn’t you have a backup?

I think it’s a grey area of responsibility too…

(is it a good thing when comments run as long as the original lengthly post?)

Your concern with your post was with your data, not so much with the service LinkedIn provides around your data, if I understand correctly.

I’m taking the engineering perspective, unfortunately, and am trying to say that in the case of LinkedIn, the two are inseparable.

All I’m trying to say (and I don’t think I was that clear before), was that the two should be separate, because while that doesn’t solve the trust issue you talk about, it makes it slightly less likely to matter.

The companies providing web apps (such as the LinkedIn website) could die like mayflies, your data would still be “safe” on the servers provided by people specialized in storing data reliably. You can trust those web app providers, mostly because they can’t mess up your data.

On the other hand, you place that trust into people who work on nothing but the problem of reliably storing your data. On the one hand, that’s a single point of failure (if they go bust, all your data is gone), but on the other hand, they know what they’re doing (unlike most web app providers).

The matter of trust still remains, it just becomes a bit clearer who you can or should trust to what level, and why.

Applications can be changed pretty quickly and easily regardless of whether it’s online or offline. The ultimate challenge would be if you’re using a Google App, storing your information on their server (securely with no concern there), but what happens when they pull the service, or your data gets lost, or they get bought and no longer provide the app or access to your data? Sure, not likely to happen to Google (anything is possible tho), however, what about small companies?

Given the immediate termination without notice clauses that we all agree to with any online service, there is the possibility that such a thing could take place. At any time. We trust them with our information, yet that information could be gone right now….

A web based office suite built on services like Amazon’s S3 might work out better, I suppose, but then that’s not free anymore. You’re charged a service fee (for S3), which isn’t even very high. But compare that to how much you pay once for a downloadable office suite, and it’s rather expensive.

It might be that enough of these services are developed in the mid to long term to make very cheap and very reliable online storage feasible, but that’ll be too late for the web 2.0 hype to still be in effect.

Let me amend my statement above: this generation of web apps won’t fly, and the next… I still doubt it.