This page is meant to brainstorm the 100 Days Action Plan for maemo.org. Please remember that it's about the maemo.org community, not the Nokia Internet Tablets.

  • Please login before making any changes. Thank you.
  • Please keep things on-topic.
  • Hardware requests are entirely out-of-scope and will be removed by community members trying to keep this page focused and on-topic.
  • Software requests which would be trivial for a third party to provide — or are already on the roadmap — are out-of-scope and will be removed by community members trying to keep this page focused and on-topic.
  • Don't put things which aren't feasible in 3 months. For long term suggestions, consider adding them at maemo.org 2010.
  • Discussion of the 100 Days agenda should be held in the discussion page.


Get the server and bandwidth infrastructure up to speed

Although things have improved some (particularly after the OS2008 release), they don't seem to be keeping up well with the ever increasing numbers of tablet users, and downtime or slowness from the critical services offered by maemo.org is unacceptable.

  • Mirror repository.maemo.org to ensure good uptime and fast response.
  • Mirror tablets-dev.nokia.com to ensure good uptime and fast response.
  • Improve the hardware and bandwidth availability behind *.maemo.org to ensure good uptime and fast response—maemo.org is too slow!
This work is ongoing and the results should be noticed way before the end of the 100 Days.--qgil 06:24, 2 June 2008 (UTC)

Increase openness

Openness is a significant part of what makes maemo so strong, putting together a plan for addressing closed-source components (particularly those directly controlled by Nokia (e.g. low-level stuff like mce and dsme, and user-space stuff like tablet-browser and the task/statusbar applets).

Perhaps Map openness would be a more accurate description, specially for the 100 Days. Nokia has not the goal of shipping a 100% open source platform. Open source is the recommended approach, but closed source is used when it offers a differentiator, an advantage over competitors. But it is good to get a common understanding on why Nokia is shipping component X as closed source, and answer when possible to requests about opensourcing something.--qgil 06:46, 2 June 2008 (UTC)

The plan

  • Identify all components (a module name isn't enough). Be like GNOME's release set list.
  • Identify all closed components in the OS at each layer:
    • Initfs/direct hardware access
    • Firmware used by kernel modules (Wifi, Bluetooth)
    • Low-level system daemons (mce, dsme)
    • User-space applications (tablet-browser, applets)
  • For each package:
    • Outline purpose in a dedicated wiki page.
    • Describe the closed-source rationale.
      • Will need some form of mediawiki template for such pages --jaffa 22:26, 29 May 2008 (UTC)
  • Document the process for working towards opening the component, or its specification.
  • Ensuring any new closed packages get a wiki page containing their rationale.
Yes, we can do or at least start something along these lines in the 100 Days. What about getting into details about this plan in a page apart, to keeop the main page clean? (same would apply to the rest of actions taken) Let's also see to the potential starting points i.e. the maemo architecture and the list of packages.--qgil 06:46, 2 June 2008 (UTC)

Get official statement about Nokia's stance towards software freedom

Attempts to make Maemo completely free software appear to be restricted by company guidelines, rules or policies. Still there is no official document describing the relationship in detail. Parts of the community want to have this information. 100 days is a reasonable timeframe to set up an official document answering the following question and provide a rationale for it:

  • Is Maemo's goal to be 100% free software (one day)?

This document could then be put onto maemo.org webserver to allow every (prospective) community member to decide on her/his own whether this is acceptable for her/him or not.

Software distribution

The website

  • Implement some sort of automated single-click push-to-Downloads for Garage projects.
Interesting idea. What about the feasibility? Niels? Ferenc? In any case integrating Garage with Downloads make a lot of sense, and the question is whether this would fit in the 100 Days or later.--qgil 06:59, 2 June 2008 (UTC)
I think we need to prevent (the need for) single file download as much as possible, but rather provide a way of uploading those packages into the extras repository directly from garage. This way people can also download the applications from the Application Manager. --xfade 13:20, 2 June 2008 (UTC)

Internet Tablet Talk Syndication

  • Provide an automatic way to syndicate applications to the Internet Tablet Talk Software Section (itTSS).
  • Each application (version) that is syndicated on Internet Tablet Talk, starts a new thread in the forums so end-users get notified of new apps as well as provide a way to give feedback to the developers. Developers themselves can also join in the discussion. This, hopefully, will help to better the quality of applications.
    • What would be the point of most the garage facilities then? I am not convinced that developers will all go looking at ITT for feedback (some projects work like that, but most rely on mailing lists and irc for example). If there was to a ITT thread for user feedback then maybe add to it a big fat 'report a bug on package x' or 'contact developers of package x'. --trickie 12:17, 1 June 2008 (UTC)
  • Provide a way for developers to easily add a bug at the application's Garage page for confirmed bugs reported in the discussion thread.
We have agreed with ITt on the general idea. It's something urgent and it would be definitely good to see it in the 100 Days. The specific plan to be discussed in an open page, either here or in ITt or both. In any case, it's clear that the feature is "syndication". Then ITt or whoever can benefit from that, and are the developers of each piece of software that decide if they want to direct the user feedback to ITt, their garage project, the project website, etc.--qgil 06:59, 2 June 2008 (UTC)
A solution could be to add a bugtracker field to the entry in the application catalog. This could then point to the developer's own bugtracker, the maemo bugtracker or a thread at ITt. This could be done on a short term and be a temporary solution until we implement a global bugtracker. --xfade 13:23, 2 June 2008 (UTC)

maemo "Application Store"

  • Actively hunt on the Net for maemo apps not yet added to Maemo Extras, talk to developers, ask and help them to commit applications to Maemo Extras. This especially applies to app porters currently active at iTT.
  • Use gronmayer's scripts to create a web site that merges applications from all known repositories into a single list and lets you browse them with MicroB using HTML UI similar to N-Gage, Apple Store, etc. While it sounds ambitious, it is not difficult to do, as we always have app descriptions and icons (form .deb files) and we also have screenshots for apps hosted at Garage. Reformatting this data in a format that can be nicely presented in tablet browser is not difficult.
  • Effectively replace "itTSS" with this website.
  • Actively convince maintainers of other repositories to join the maemo extras repository. This would save some headache from offline and broken repositories.
  • Make sure that all applications which are actually available in the repositories can be found in the Maemo.org downloads section.
From a Nokia point of view the interest is to have a single repository for distributing third party applications. External repositories should be an exception targeting very specific needs and users. Nokia is not interested in blurring the boundaries between a community supported extras repository and the repo someone has setup somewhere.--qgil 06:59, 2 June 2008 (UTC)
While it is obviously better to have a single repository, both for users and for Nokia, realistically it is not going to happen soon. So, there is still a place for a repo aggregator like Gronmayer to exist. Additionally, creating an AppInstaller-like AJAX web site for Gronmayer specifically targeted for displaying in MicroB will let you test different AppInstaller designs before implementing them in a real application and pushing them to the users. It is really easy to change a piece of PHP code but changing a native app and updating everybody's tablets with it is more dofficult. fms
If you want to have any specific action around the Gronmayer aggregator & tools, that's totally fine. Go ahead. I'm just saying that Nokia prefers to invest its time and resources fixing the root of the problem, even if it will take longer than 100 Days.--qgil 07:59, 2 June 2008 (UTC)

The repositories

  • Again, because this can't be stressed enough, get the infrastructure behind the repositories up to speed—we need servers, we need bandwidth and we need mirrors.
This is already being worked on at the moment and we hope to have results soon. --xfade 13:31, 2 June 2008 (UTC)
  • Streamline the acceptance system for Extras(-devel), (not necessarily easier or less stringent, but more straightforward and clear).
Please try to point out what is not clear or what is confusing. I am trying to get a clear image of the problem, so I can fix this. --xfade 13:31, 2 June 2008 (UTC)
  • Lay out the groundwork for a peer-review system for Extras acceptance (or devel to Extras promotion) to help ensure good quality assurance on its packages.
Maybe we need to start a separate wiki page with a proposal for this. This is not something we can do in a few lines here. But I would be very interested to work on this. --xfade 13:31, 2 June 2008 (UTC)

Developer documentation

Clean out the cobwebs, remove the cruft

  • Get what's still useful and mostly relevant completely up to speed for maemo 4.x.
  • Archive the outdated information away from the maemo 4.x stuff.
  • Mark the outdated stuff clearly.
Agreed. What about a new wiki page listing what is really essential and should be put upfront? One approach could be to have the essential bits well written and presented in static-ish CMS pages and stable PDFs and then move all the rest to this wiki ecosystem. As soon as new wiki content gets really good we could productize it and promote it.--qgil 07:34, 2 June 2008 (UTC)

Get organized

  • Put together an easy-to-navigate, sensible index for the documentation content.
    • Place references/howtos/tutorials onto one page so that developer does not need to click through several menus.
  • Improve the search (google?).
The plan is to have all the official HowTos and docs under a well structured Maemo Reference Guide. Deadline: Diablo release. This reference guide would sit between the Quickstart Guide and the Training Materials. There were some fixes in the search and should be working much better now.--qgil 07:34, 2 June 2008 (UTC)
The website search has already been improved, will add this wiki to the search too. --xfade 13:37, 2 June 2008 (UTC)

Get focused

  • Add porting FAQ wiki page detailing common problems developers will run into (i.e. application is killed 3 seconds after launch) and how to deal with them. Provide examples of typical GTK/Motif/etc. application changes needed to properly Hildonize the application. List ways to deal with porting of toolkit-specific functionality whether it be internationalization or mouse/keyboard input.
  • Make it possible for logged-in developers to annotate any place in the documentation. Link to annotations from documentation.
  • Provide an example of simple build environment *not* relying on AutoConf and its friends. A single includable makefile should suffice, when used with SB2.
  • Clearly *say* in the SB readme that it is not possible to debug every application on the desktop, show how to test applications on the target device using SSH/SCP or some other means.
  • Clearly define what changes is made by default on gtk : like GtkTreeView with hidden header columns by default, or image-button off ...
The idea is to have the sources of the official documentation here in this wiki. Users could add comments in the discussion pages, perhaps some users could get edit rights too. To be decided. The Nokia teams would updated their part of the documentation here and then all would be revised and conveniently package in PDF when a new stable release comes. So yes, we can put this in the 100 Days and then let's see how far we go in this timeframe.--qgil 07:34, 2 June 2008 (UTC)

Look towards the future

  • Define types of applications that will be useful on the Internet Tablet
  • Stress the fact that the Internet Tablet is not a PC and apps should be created/ported with the tablet form in mind. Don't just do a direct port of an existing app. Aim for quality and Internet Tablet usability.
  • Focused discussion/guide on User Interface so apps will have a consistent look as well as provide a similar way to interface with the user
  • Maybe provide a few simple stylesheets and JavaScript libs for creating quick iPhone-like web apps running in MicroB. This should be very light, very easy to use, and targeted to casual users.
Good ideas, they can be pushed by Nokia and the community. The point about Javascript libs is imho more of a 2010 Agenda thing, though. I don't mean we wouldn't have anything before 2010, but 100 Days is just too tight. Besides, the work to be done goes beyond maemo.org.--qgil 07:34, 2 June 2008 (UTC)


  • Re-write and expand Introduction to better serve as a useful introduction for newcomers to the platform.
    • Intro section should include "Who is the maemo community?, What is the maemo platform? The maemo software architecture, How does maemo.org work? Quick start guide to develop on maemo software, and presentations".
  • Single sign on for maemo.org/garage.maemo.org/wiki/bugzilla -> would make community participation easier and the 'karma' calculation (if needed) too.
Ferenc is working on bugzilla, IIRC. The others already have one account for all. --xfade 13:49, 2 June 2008 (UTC)
  • Website information does not suit the needs of newcomers: How to install or do xy on the IT? Where is recent information about OS2008? Is this page, which I am looking at, an outdated or a recent page on maemo.org? What resources are available for me (alias I am confused by unconnected information sources: gronmayer, internettablettalk, planet.maemo, official Nokia site, internettabletschool, maemo ...) Where is a detailed roadmap for Maemo? Is there an application wishlist for OS2008? Where can I give input/ideas as end-user?
All good points, agreed.--qgil 07:38, 2 June 2008 (UTC)
The whole intro section should to be redesigned and turned into a proper starting point for new visitors. --xfade 13:49, 2 June 2008 (UTC)
  • planet-maemo in other languages: For me planet maemo is a very nice aggregation service. This should be available in other languages too. Maybe there should be two versions (per language) of planet maemo: one for developers and one for end users.
We can do multiple languages by creating a folder for each language and adding the proper feeds to those folders. There is also a possibility to display news based on what you have thumbed up and down in the past, but that would be something that can only be done after the planned server upgrade. --xfade 13:49, 2 June 2008 (UTC)

Style and format

  • Maemo.org can benefit from some face lifting - right now on 1280x1024 (this resolution seems to be very common to developers) only half of the real screen estate is used. With some loose block positioning both 800x480 and bigger resolutions can be supported. Maybe even specialized tablet finger-friendly look for those preferring it.
    • Agreed, we have a lot of size and font problems that need to be addressed. Some example stylesheets would be welcome though. --xfade 13:49, 2 June 2008 (UTC)
    • Resolution seems fine to me at the moment, and it works fine on the tablet without requiring the extra effort of maintaining a separate tablet style. Perhaps reducing the min-width to not require horizontal scrolling with the tablet browser windowed would be useful, though. generalantilles 21:02, 29 May 2008 (UTC)
      • There won't be any extra effort on the resolution maintainment - this can be done with minor CSS modifications. As for the separate tablet style - this really requires extra effort and if done should be entirely optional. bundyo 21:26, 29 May 2008 (UTC)
  • More relevant information displayed on front page, preferably customizable blocks like Netvibes and iGoogle. Since the content is gzipped, that won't be too harming to the traffic. Blocks can be optionally auto updated for those that like to keep their browser pages open (and if Prizm is ported - even in the tray). For instance, a "new bugs" section with voting on the fly will boost bugzilla usage.
  • Make links more standardized. New users often expect links would be underlined, even subtly, or appear to be a button of some kind. Don't rely on hover effects because the tablet can't practically use them. See Wikipedia on 'Mystery Meat Navigation'
    • This is much less of an issue for a lot of people —generalantilles 19:49, 31 May 2008 (UTC)
    • Please stop expecting your advanced experience of new users. Read up on web accessibility, be inviting to the new users.--boxofsnoo 23:08, 31 May 2008 (UTC)
      • I really don't see the issue (and never particularly have with this much-abused "Mystery Meat" nonsense), and I don't think it has anything to do with my experience, wikipedia (your source) uses the exact same system for link identification. . . . Some specific examples would probably help your case. Anyway, this should probably be taken to the talk page. —generalantilles 23:20, 31 May 2008 (UTC)
  • Format and style need to be unified across as much of the site as possible (excluding things like Garage and Bugzilla). Take, for instance Planet and News, two pages that should be very similar, if not the same. Perhaps take News' style and format and apply it to Planet (add the contributor's avatar to the upper right of each article? Much like slashdot does with their article category images.), as the News style seems to offer a cleaner look that better utilizes the available space.
  • For the outliers like Garage and Bugzilla, at least the style should largely be unified with the main site—using the same fonts, same colors, etc.
OK to have a well integrated, efficient and nice-looking layout across maemo.org. Ok to discuss in detail in a page apart and ok to have a basic plan agreed in 100 Days. The execution will take longer, though.--qgil 07:38, 2 June 2008 (UTC)


  • Develop a recommended usage policy for garage.maemo projects, taking into account turning off GForge modules not needed (perhaps defaulting in a subset rather than all, for new projects). Further discussions on whether non-core bugs should be in central Bugzilla and garage trackers merged/closed.
  • Close misnamed (and now misused) maemo2midgard-discuss mailing list and create maemo-web alongside existing mailing lists for overall discussions about maemo.org sites.
Both points agreed.--qgil 07:45, 2 June 2008 (UTC)

  • Today we have too many channels (ITT, maemo.org...). Maybe have some more focus like: maemo.org for developers and ITT for end users and something that links them so software releases can be announced automatically on ITT and users from ITT can easily post bugs on garage bugzilla for example.
    • Perhaps develop a plan, but this isn't specific enough to achieve in 3 months, IMHO --jaffa 22:10, 29 May 2008 (UTC)
      • I'm not convinced that two channels is too many, either. —generalantilles 20:04, 31 May 2008 (UTC)
  • It's unclear where to report problems about packages found in the application catalog. A single bug tracker is needed, or at least a catalog that would redirect the user to the appropriate tracker from a common start page. Or we could set up a "maemo" distribution on launchpad.net, which would let us integrate with the bug trackers of individual packages.
Nokia is open to discuss any possibility of integration of maemo.org with third parties i.e. ITt and also using any external service i.e. launchpad.net. We can commit to overall plans for the 100 Days exercise but not on execution. Having a better and common understabnding of how this steps would fall in the 2010 Agenda would definitely help to see what are the right first steps.--qgil 07:45, 2 June 2008 (UTC)

What the community needs to do

  • Put together a plan of action for moving forward with a community-maintained Hacker Edition (based on what Quim said here)
Having a specific proposal agreed by the community would definitely help.--qgil 07:50, 2 June 2008 (UTC)
  • Put the pressure on developers:
    • Encourage developers not using a repository to package their applications and push them to Extras(-devel), and get developers with 3rd-party repositories to close them down and push their stuff into Extras(-devel).
    • Encourage developers to follow proper packaging guidelines (based on the draft here).
    • Encourage developers launching applications on Downloads to include as much information about the application as possible (screenshots, good descriptions, good changelogs).
The aim is good and we support these ideas. It would be good to define the specific actions to be done in these 100 Days, though.--qgil 07:50, 2 June 2008 (UTC)

Update developer libraries

  • gcc-4.x, glib, powervr, it's important to give developers much more time to play with newer tools than short before a major upgrade takes place (of course with disclaimer that nothing is guaranteed to be shipped in a certain way)
  • Update the vmware appliance with sb2 and python2.5 setted. And more publicity on the vmware appliance, as there is already one on garage, but many don't know it.
This is out of scope in this maemo.org exercise. About giving more publicity to https://garage.maemo.org/projects/maemo-sdk/ and http://maemo-sdk.garage.maemo.org/ - it's currently an alpha and the developers don't feel like advertising it much more. But anybody can follow the Garage project and, well, nobody is stopping you from advertise it more.  :) --qgil 07:53, 2 June 2008 (UTC)