That's what we're not keen about. (replace 770 users with N8x0 users...)


Well, that objective is still clearly rather vague "work in progress"...

So there's no current plan/hope to make Maemo 5 (Fremantle) run on the current N8x0-class devices? Just seeing the odd bug fix percolate down from the next OS upgrade to current devices sounds indeed wasteful when the userbase is clamouring for a proper upgrade (even if it's "Fremantle Basic" without some of the more advanced goodies and eye candy).

So to answer your question with a counter-question: What if investing heavily in making future releases (at least +1) available to current users and devices helps make the platform attractive and thus sustainable?

I.E. I wonder if it's sustainable to split the userbase and developer community to N8x0/Maemo 4 legacy and N9xx/Maemo 5 groups.

Only when the Fremantle SDK ships and eventually when Nokia starts handing out new "loaner" N9xx hardware to developers we'll find out how self-sustainable the legacy N8x0-platform really is. Not forgetting that some developers would also rather play with alternative platform ports (Debian, KDE etc.) on their "legacy" devices...

Looking back, Maemo 4 (Chinook) arrived in the scene in late 2007 and it has basically had one minor update to 4.1 (Diablo) in mid-2008 plus associated 4.1.1 bug fix SSUs.

From mid-2008 forward, and until Fremantle with new N9xx hardware expected around mid-2009, what are the N8x0-users (i.e. most of the Nokia Internet Tablet userbase) looking forward to in this very fast-moving scene?

I don't know how much my personal views represent the average N8x0 userbase, but the longer my current device continues getting meaningful updates the likelier I am to consider buying another NIT. (Esp. if there was still some compatibility/synergy left between the old and the new device...)

To be fair, this has only relatively recently been decided, and if there's consensus on an even better process, nothing should stop it from being adopted. There's also inconsistency: have all the WONTFIXed ones been corrected to "FIXED (Fremantle)" yet?

Having said that, some reaons why leaving it left open isn't really very viable with the tools:

• It's not easy to see which requests have been dealt with, and which are newly opened.
• It's slightly unrealistic: in the current climate no-one but Nokia really are going to provide these fixes to earlier OSes, especially if the Fremantle fix is to rearchitect a portion of the system to make meeting the requirement easier.

Perhaps a modification of the "target" milestone would help. Ideally, of course, the issue would be assigned to the appropriate designer/developer/reviewer/architect to act as a master for the development. It would then be updated with specs and assigned through the development workflow, finally being FIXED when the code is committed to some repository somewhere (hopefully a publicly visible one).

Failing that, what about keeping them open and assigned to "fremantle@maemo.bugs", again until some (internal) build of the software meets the requirement as outlined.

Sure, if there's consensus on a better process, but there's clearly no consensus and I've yet to see a better argument for changing the current process than "because we're spiteful about Fremantle and don't want to face reality".

Now, you might say that enhancements should match bugs in being marked FIXED only when the code has been committed, but enhancements are different beasts, and their scope is frequently significantly larger than the typical bug, so tracking that "all appropriate code has been committed" point is much more difficult (and not a particularly useful thing to pursue).

So, enhancements are closed when they're roadmapped, and bugs are closed when the code is committed. If you can't deal with bugzilla not differentiating between hardware releases then I recommend unsubscribing from your bugmail and moving on.

I've corrected the ones I've run across (maybe a half-dozen so far), so there's probably not a whole lot of them left. Either way, the inconsistency stems from incomplete or changing information (i.e., the feature was roadmapped after the bug was resolved, whoever resolved it didn't know that the feature had been roadmapped, etc.) than an incomplete policy changeover (i.e., the person resolving knew it was planned for Fremantle but closed it WONTFIX for the current release), so it's not really relevant to this particular discussion.

I think you really have a problem with people having different opinion than yours, you always think they don't understand what they are talking about. Which makes you look really intolerent.
You should work on that.

I don't really feel the need to comment on the topic of this thread since "It's the way it works" and I feel like I lost my time.

Yes, they state that they believe the way things are done should be changed and you reply "But that's the way things are done." IMHO that doesn't add much to the conversation, and when repeated over and over comes off as rather smug.

Feel free to convince me of your point of view.

If you have a cogent argument for why your method is better, then make it. So far, restating your opinion over and over and backing it up with emotional appeals hasn't been very effective.

I don't get why everyone's panties are up in a bunch over this. The boys at Nokie tossed the '07 and '08 hackers out and I'm pretty happy with em' (running off a card with the faster kernel, of course) on my 770. If they do the same with Fremantle (with all these bugs fixed) and the N8x0s, I'll be a pretty happy camper.

I don't feel like Nokia abandoned my 770, and I don't think they'll abandon the N8x0s. We'll get those bug fixes. Eventually...

Although, considering I just paid about $400 for this N810, I'd appreciate a fully supported Maemo 5 on it. (And I could have it, if someone would pony up for those nasty drivers we're missing...)

Indeed.

Yes, exactly (and within a reasonable time frame).

We have been told repeatedly that Nokia has learned from the 770 fiasco and such mistakes won't be repeated. We were also told that the N8*0 devices will be supported by at least Diablo and Elephanta. Some of us went out and bought N810s based (at least in part) on the above.

We're now in a situation where:

• All purchased N810s (not to mention N810WEs!) are less than a year old right now.
• Elephanta has been dropped.
• We don't know whether Fremantle will be supported on N8*0.
• Even if it is, it's at least 6 months away.
• Existing Chinook/Diablo bugs get closed WONTFIX because Nokia doesn't have the resources to deal with them.

This is unfortunate and has considerable dampening effects:

• effectively killing any motivation to continue reporting/testing Diablo bugs and
• creating reasonable cause for concern that the next generation device could also end up being unsupported after a few months.

I hope this is just a communication problem but until/unless Nokia announces definitively that Fremantle will work on current hardware people will continue feeling frustrated and demotivated.

Don't you think that no matter how upset they were at the time, most 770 users welcomed the new software and hardware since they marked clear steps forward? Fremantle is a bigger step forward and... well, we can continue this discussion when a new device with Maemo inside is announced.

If you do the math, we are talking about a process of several months between the first and final Fremantle releases. For what we know already today, Fremantle is much more open than Mistral/Gregale were at their time. Also, the Maemo community is now much wider, with a deeper knowledge of the platform, with some dedicated members and several instruments of collaboration and decision. No matter what Maemo SW does with Diablo and Fremantle, the situation is radically different than in January 2007.

Perhaps the Maemo Variants objective sounds vague for many of the community members that actually would benfit from them. Perhaps we could work more on them instead of playing tennis in the forums. It's a wiki and you are welcome to get into details as much as you will. From the Maemo SW side is not vague at all: implies certain infrastructure and support, opening and relicensing components, and we are working in all this right now.

About the rest of your points and many of the points made by others in this discussion:

- When the first Fremantle release comes we can discuss on top of real code (very soon). Before that we are going to probably repeat ourselves from our current positions.

- When the Fremantle beta release comes we can discuss on top of a full feature description that will allow you to understand the amount of work Fremantle requires. We can then discuss how feasible/useful would have been to backport those features and fixes to Diablo as well.

- When a new device with Maemo inside is announced we can revisit all this discussion since then everybody will have their opinions based on public knowledge. Now you have to base your arguments in assumptions, which is not the best foundation for a fruitful debate.

I don't have a method, I don't give a flying **** if Freemantle bug-fixes are made available for the N8x0, as I have a 770. However, your manner is irritating and unpleasant. Feeling the need to add a post essentially saying "That's the way it is." in response to someone suggesting an alternate way of doing a thing is rather hostile.

I find reading this message board rather enjoyable, but when I get to one of your "I'm looking down at you because I think I'm smarter than you" posts, it takes away from that enjoyment. Just changing your picture to one of you smiling isn't going to change the perception one gets of you from your manner.