I did it for you.

Except that the formula is not JUST working for the iPhone. Android did the same thing, even having contests to load up their app store (the Android Market) with interesting content before the phones even shipped.

And, application ecosystems do seem have a huge impact on any platforms success. You could argue that's a large part of what kept Windows ahead of everything else, even before MS could legitimately throw it's weight around like a monopoly.

Though, to some extent, I think the Application Manager is fairly close to being an Application Market/Store already. It just doesn't have a provision for paid applications (well, once you pay for directly through the market, as opposed to paying later through the individual applications).

Android is in this respect is much closer to iPhoneOS than Maemo or Linux in general - why do you think they chose the route of managed code ?

Nobody said otherwise. I just don't believe that the iPhone ecosystem model is the ONLY possible successful ecosystem model. If you used the windows example, you see that their ecosystem is very different (and as much as I hate to admit it, it does work, too).

The similarity is very superficial, basically all the similarities end right at the ability to retrieve applications from a common place (which might seem a big deal, but is just the tip of the iceberg ecosystem-wise). You don't even know the LICENSE of what you download (just that it's not of Nokia origin). And that one is a WONTFIX for fremantle right there. And not mention all the apps (can) have root abilities. So no, app manager will not be AppStore, at least not any more than ANY desktop distro's package management tool will.

Whether or not Android as an OS is closer to the iPhone than Maemo is irrelevant. An App store's value is a single tool presented to the user from which they can search for, and find, and manage the installation of, useful applications.

In that regard, the Android Market, the iPhone App Store, the Pre Application Catalog, Maemo's Application Manager, or even Ubuntu's "add/remove" software mechanism, are all in the same niche.

Their ecosystem is different, but the example is that an ecosystem plays a huge role in success. The windows ecosystem predates the popularization of the net, so clearly they built their ecosystem in a different manner. But, the ecosystem is probably a bigger factor to the success of the platform than than the quality of the platform itself (again, look at Windows). What a solid application manager provides is a way to quickly build that ecosystem by letting developers quickly and easily connect with their customers, and visa-versa. In essence, it bootstraps and accelerates the growth of the ecosystem. And that's the connection/point between an Application Store and the success of Maemo.

But it's really the main value that an App store provides. A single place for the users to go, a single presence that the developers need to focus on.

I don't think that's a gating factor for an App Store. I don't know the LICENSE of the apps I install from the Android Market. In fact, I don't think I've seen a single EULA, nor other license statement, from the apps on the Android Market.

The differences between what the Application Manager does, and what is important in an App Store, are, IMO, minor. Could it be better? yes. Does it need to be dramatically different? no. Do they provide the same basic service, and same value add to the user community and application ecosystem? yes.

What are those changes? IMO:

1) better descriptions of the apps (most of them are pitifully brief, and often provide no useful information whatsoever ... I mean, I've seen apps where the description is nothing more than "Blah for Maemo" or something equally inane), including license information.

2) a mechanism for charging for non-free apps, and/or informing them about whether it's totally free and open, just free, charges for support but not use, charges for use, etc.

3) a standardized mechanism for informing the user what security issues the app has (accesses the internet, accesses your PIM information, accesses your IM/SIP/Email/etc. application, runs under priviledged access, etc.), and a "are you sure" part of the dialog, informing them of those items specifically.

The only one of those that would really quire a significant change to Application Manager is #2. #1 and #3 could be worked into the under-utilized package descriptions (though, #3 would probably be better done through flags than textual free form data).

For monetary safety reasons, I would suggest that the only repo that was allowed to do #2 would be an official repo, managed by Nokia, for 3rd party apps. But all of the repos should require conformance with #1 and #3.

...which brings to my mind: why are the Maemo Application Manager and Maemo Downloads so separate entities? An improved application manager would utilize all the cool information Downloads has, like ratings, comments, screenshots, "hot apps" and so forth. A much more integrated and pleasant experience than the current quite pedestrian list of .deb packages...

And best of all, the packages listed would mostly be Free Software, not the "modern shareware" App Store has (and Ovi Store probably will have).

You're missing the main point of my post. Most linux distros (Maemo incl) are NOT RUNNING MANAGED CODE. This means that your data and system are at the mercy of the packages. How are you going to prevent an app having root access not messing with your maemostore account or point 2) ? Who is going to guarantee you (and how) that the point 3) is declared correctly (and respected) ?

The bottom line is that people see this from a user perspective and think AppStore (and thus ApplicationManager) *IS* the ecosystem - but it isn't, it's just a facet, the distribution arm. That cannot be transplanted to other platforms until there is a clear mechanism/enforceable policy on the system level to separate the applications from the OS and other apps (iPhoneOS has this sandboxing, Android too). Maemo doesn't do this, and until it does, all the talk about a payment system, better descriptions, etc are superfluous.

BZZZZZ, wrong answer

(a not so happy, iMac induced iTunes user).

it's a bit sad that this very interesting segment of discussion about app managers and stores will be lost around the page 122 of a thread with an unrelated tipic and origin... If a moderator could take these posts out in a new thread it would be easier to gather the attention of more readers, perhaps including those thinking how to improve the developer offering and distribution channel in Maemo.

"If a moderator could take these posts out in a new thread"

I was hoping that this would become the "thread that never ended," something like "the man who never returned," of ancient folk song fame.

("he will ride forever 'neath the streets of Boston, he's the man who never returned")

Time for a 'weekly maemo newsletter'? Could be funded for an individual to monitor mailing-lists, forums, bugzilla e.t.c. For me that seems like a perfect solution (ala Zac Brown and LKML). Oh, Quim, mail me, I have the perfect guy.

A lot of the contention lately stems from a lack of knowledge. A lack of knowledge of bugzilla, iTT, IRC, the mailing lists e.t.c or something equally important. Somebody to highlight the issues would be huge asset to Nokia at the moment, there really is a missing gap there.