I am a believer in updates and progress, so I use CSSU, but the woods are full of people who say "if it ain't broke don't fix it."

Can't agree. We're not ISO, so no higher authority decide which one is latest upstream. If, for majority of users and developers CSSU is latest upstream, it is the case. Howg. Of course, it's FOSS, so if You like to, you can write programs for PR<whatever>, or even make them incompatible with CSSU on purpose, and no one should complain. Yet, i would agree, that if it's found that some program is incompatible with PR1.3.1, PR1.3 or any other version, it's nice practice to write about it in description.

Fortunately (from my point of view) CSSU is considered "mainstream" - at least after Stable version was released (yet, for many - including myself - it was stable quality for a long time), so it's desired target for developers, bigger or smaller (or ultra-smallish-tiny ones like me).

I understand problem with contacting maintainers of long non-developed programs very well. Yet, it's FOSS again, so most of the times, you don't even need to contact maintainer - fix code yourself, and you can upload packages to extras-devel without being maintainer. This was the case with TrueCrypt - I requested maintainership *after* -devel version was uploaded successfully. I was lucky, cause long-inactive original developer still had his mail active, and approved maintainership request, so I was able to promote it to extras-testing

in other case, I would be doomed to wait 2423543543 unspecified time cells, until X-Fade a) gets contactable by someone else than merlin1991 b) actually willing to perform some work that he's paid for c) wouldn't lie to me that more than one maintainer of package isn't possible. which (as situation) is IMO real problem, and even CSSU is hurt from that, actually - we (the Community) started to working on fixing it lately.

Of course everyone is free to have any opinion on any subject, but I think that we should focus on real problems, not on creating them from scratch

/Estel

So, can someone list all the fixes to the actual N900 (excluding anything to do with portrait mode).

Nokia left us a bug list with a sh*tload of "WONT FIXES'. From what some peeps here are saying, they're all being fixed.

I don't use CSSU, simply because it doesn't offer anything above the current Nokia Kernel that I can use.

cssu project seems a little...frozen atm

sadly

You miss the point. Creating incompatibility with a one-liner 'fix' (that is actually a remap so far down the line) is a major issue. At least when you end up on the receiving end, like myself. It wasn't working in certain applications (screen+vim/less), after 'fix' it isn't working in other, not as claimed in the changelog. If you happen to use screen + x app, you're happy (not sure what for considering you can open as many xterm windows as you like and manage them easily through task switcher, but oh well). With KP_Enter being around here since at least Diablo a decision to remap shouldn't have been taken lightly, yet it was. That is just one example of lack of proper QA/UAT testing.
As to versions, CSSU might be TMOfficial, doesn't matter really. 1.3 was supposed to be last, if another security, or other update will come, I am betting it will be 1.3.2, not 1.3.3.8, but hey, you're free to fork and divide as long as there are at least two devs left (sorry, but again I feel this is exactly what should be avoided at all costs with shrinking user/dev base):

Well, sometimes it is a good idea to at least look at commit log on gitorious before making such а conclusion. What is "frozen" is not a project, it is CSSU-T maintainership. Because of the way CSSU project was set up in its early days and thanks to MohammadAG not willing to give up on being maintainer of the CSSU-T we have new version ready (but not distributed) since the end of November or so. And it was that way all the time for the last half an year or more - every time it was a battle to find MohammadAG and to convince him to push the new version. It was his laptop cooling fan faulted, it was his internet connection in uni, it was his n950 at customs, etc. etc. And if someone does not believe in my words, he can ask arcean,joerg_rw, merlin1991 or someone else who is active on #maemo-ssu for confirmation.

@szopin - You have a point saying that all you see about CSSU is portrait mode. But AIUI it is the main "selling point" of CSSU, as Fremantle portrait mode was the most demanded feature during the last 2 years. On the other hand it is not all about portrait mode at all. Maybe you are right in sayng that you cannot see any bugfixes in CSSU changelog, but it does not mean there are no such. Examples - hildon-desktop 100% CPU usage, missing GPS precision in pictures taken, recorded videos stutter (should be fixed in the update yet to come). Still having my coffee so those are the only ones I can recall, but there are more, trust me on that (tm). modest ,qt(for sure) and pdf-reader are upstream versions AFAIK. etc.

The plans include community kernel, which will allow us to fix broken omap thumb support and thumb re-compile most of the system libraries. That means a pile of additional free RAM. Community kernel (based on kernel-power of course) will allow us to run devices with SmartReflex turned on by default, increasing battery life almost twice (compared to stock Nokia kernel). etc.

Without wanting to repeat some good comments made in this thread, I'll just summarize why I don't use CSSU:

1. My N900 running PR1.3.1 with KP47 is essentially bug free, so no direct need for any updates.
2. I don't care for portrait mode.
3. AFAIK with CSSU modest crashes frequently. That's an absolute no-go for me until I get confirmation that this issue is fixed.
4. My uptime is usually counted in weeks, if not months. I only reboot when I want to reboot, which is almost never.

I realize that CSSU brings some benefits (such as moving to more recent messages when deleting an e-mail, I hate the standard behaviour of Modest in that respect), but if it's going to crash anyway, then I prefer the annyoing, stable, behaviour of 1.3.1 modest.

I am sure there are, but for all my usage scenarios KP is enough it seems. No problems with hildon, stuttering, modest (thoujgh it did crash on me thrice soon after installing kp). It was just my opinion(tm) why I don't use CSSU. Inconsistency between KP_Enter in SDL vs Return in console is just one of the omitted issues by all testers it seems. I had a problem on vacation soon after deciding to test CSSU which left me with no pictures, tough luck. Definitely didn't want to spend hours here looking for solution instead of enjoying myself, but it left a bitter taste (pretty sure it's fixed now, as it was almost half a year ago). Don't really find portrait useful, so with abovementioned inconsistency/incompatibility and no issues with KP I see no point. Might give it another run, though not on the dev device for sure

modest is fixed and is rock-stable (it was crashing because it was thumb-compiled all the time ).

I'm yet another one of these non-CSSU user with a perfectly running N900 still on KP 42 @ 850MHz.
It's my primary phone and most things I do are terminal-related.
My uptimes are usually 1 or 2 months but that is not limited by Maemo's long-time stability but by ID10T errors like forgetting to recharge it or not properly closing the battery door after changing an SD card.