Tuesday 21 December 2010

Captain's Log: Stardate 21122010


We are at the end of another long year for the Vinux Project and we have managed to release Vinux 3.1 in all of its various forms before the new year. So now it is time to hang our developer's gloves up and have a good rest! So thanks to everyone who contributed to this release in whatever way they could, Vinux wouldn't be possible without a friendly, helpful community of people who all have the same goal: open-source accessibility for everyone. But as always, I am thinking about the future and trying to anticipate in which direction the project should go. So here are a few things for everyone to mull over during the holidays, which will roughly form the agenda for the online meeting I would like to hold in the new year, probably on our IRC channel.

Firstly, while the community has slowly grown over the last year, we are still painfully short of core developers who have the time and skill to keep the project going. Granted we now have lots of people who help out in a variety of peripheral ways such as; hosting vinux iso images, writing new packages, testing alpha and beta releases, answering queries on the support list etc. I am not going to name any names (in case I leave someone out) but you know who you are. However at the moment there are really only three core developers: Bill Cox, Mobeen Iqbal and myself who actually produce Vinux releases. Bill was the main driving force behind the Vinux 3.0 release, but due to health problems this year he has not been able to take a lead role in the latest release (although he still provided invaluable help and advice). Mobeen produces the VMWare Virtual Edition of Vinux and a special version of the USB Edition which can be installed to a pendrive from Windows, which are very popular with newbies who are afraid of messing up their Windows installations (with good reason). This means that I produce the final release of Vinux, create all of the different versions, upload them to the servers, update the website and then make announcements on the relevant mailing lists and forums etc. This may not sound like a lot, but given that I have a full-time teaching job and a family to look after, I can honestly say that I spend nearly all of my evenings, weekends and holidays working on Vinux to the exclusion of all other activities. Not that I am complaining or trying to be a martyr, I took on the responsibility when I started the project, but sometimes the pressure of having to work on Vinux instead of relaxing becomes overwhelming. So as always, here is my annual appeal for help!

There are lots of different ways in which people can help with the development of Vinux, irrespective of their skills and knowledge. I will now identify some roles that each and everyone of you can play in the Vinux project:

Testing - This basically means testing out alpha and beta versions of Vinux before the official release is made. Indeed many of you already provide feedback, but up to now it has been in a very haphazard way. We need to formalise the testing process with specific individuals identified to test specific packages and provide detailed feedback - not just X doesn't work. We will also need someone to coordinate the testing and feedback process, so we could do with a 'Testing Taj' to step foward...

Documentation - This is something that users of any level can do, you just have to document in detail how to use a particular package with Orca or Speakup etc. Again in the past this has also been fairly haphazard, but now that there is a specific repository for the Vinux Documentation Project this should become more coordinated, with individual users being asked to provide a guide to specific packages etc. Again we need someone to take responsibility for coordinating this, a 'Document Dictator' if you like...

Publicity - This means what it says on the can - making sure people find out about Vinux through mailing lists, forums, blogs and other websites etc, but without annoying everyone or being accused of spamming etc. The other important element to this role(s) is the need for realism. We can't make out that Vinux is the solution to everyone's computer needs - it offers advantages and disadvantages compared to other operating systems and we mustn't overblow our own trumpet as this will raise expectations, leading to inevitable disappointment for some users. Vinux isn't or probably won't ever be right for all VI computer users. So what we need is an 'Realist Reporter' to take on this role...

Mirrors - Providing mirrors on which to make Vinux available is of vital importance. Having mirrors around the world allows people to get Vinux from their nearest server, or at least provides options if one or other of the servers is down. We actually have a reasonable number of mirrors at the moment, but if you can host even one iso on server with reasonable bandwidth speeds (unlike mine) then drop us a line - we need more 'Mirror Mages'...

Website Management - This role(s) would require more technical competence than the previous two roles, as it would involve managing a website(s) and/or mailing lists etc making sure that all of the information is up to date and the iso images are available from the relevant mirrors. I would be happy to hand over this side of my responsibilities to other trusted members of the community as it would free me up to spend more time on the development side. So we need a 'Website Wizard' to step forward as well...

Package Development - There are already several people producing packages for Vinux and the wider Linux community, but this is an area were people can work relatively independently and produce packages which they think will be useful for themselves and others alike. The only communal element is uploading it to Launchpad, building the binaries and then providing technical support on the mailing list for users if required. If you have an idea for the next accessible application which will take the world by storm, then get coding. We need more 'Package Peasants'...

Distribution Development - This role probably involves the most responsibility and stress, but technically speaking it isn't as hard as you might first think. It requires a large amount of patience, determination, effort and time, but not that much intelligence - just a spattering of Bash scripting and a basic understanding of Ubuntu, the Gnome Desktop and the Accessibility software. So if you think you are a match for my determination and stupidity, then consider getting involved in this side of things. This is a role that can easily be split between different people, either taking turns to produce releases, or producing different versions of the same release. We need some 'Distribution Demons' as soon as possible...

Each if these roles could be taken on by a single individual or by a team of people, although each team would need a coordinator of some kind. So if you fancy playing a more active part in the development of Vinux please get in touch, privately if you prefer to discuss any concerns or anxieties you might have...

Now it's time to discuss the big issues which we have to face in the coming year:

The first issue is the proposed move from the Gnome Desktop to Unity in Ubuntu 11.04. This may or may not cause problems for screen-reader users. Luke Yelavich has been seconded onto the unity development team to make it accessible, which is a positive sign. But I feel Luke is always having to play catch up with decisions taken by Canonical in which accessibility isn't the highest priority. I suspect the motive behind switching to Unity is to give Ubuntu its own unique interface which will work well on netbooks and touchpads etc. So depending on how things develop we may have to adopt Unity and any limitations this has, or stay with the Gnome Desktop until Unity is usable.

Related to this is the question of whether we should stick with Ubuntu as the base for Vinux, switch to a different distro, or diversify and produce more than one version. Given the current lack of core developers, the third option seems unrealistic (unless some maverick decides to start producing their own version of Vinux) so that leaves us with the perennial 'Ubuntu or not' question. Ubuntu offers many advantages in terms of ease of use, hardware support and the size of the user community. On the other hand it has a very fast release cycle and I suspect it may have commercial motives in the long run and this may limit our freedom to release Vinux at some point in the future. There are other distros that offer similar features to Ubuntu e.g. Debian, Mint, Fedora etc, but they are also release based, meaning that the only safe way to upgrade is to install the latest version. On the other hand there are rolling release versions of Linux such as Arch, Aptosid and PCLinuxOS which can be kept permanently up to date without re-installation, but they are not easy to make accessible or remaster. The only viable option I could envisage using at the moment would be Mint Debian, which follows a rolling release based on Debian Testing, but also offers most of the features provided by Ubuntu, in addition to multimedia codecs etc.

Finally there is the issue of what features and packages we should and shouldn't include in Vinux. To some extent this is determined by the room on the iso image, but we could for example produce of lots of simple accessible tools using Zenity and Bash scripts (like Speedy-OCR) or plugins to existing applications (like Markup-Binder) that would make users live much easier - or should we encourage people to learn how to use the standard tools provided by Ubuntu or in the repositories?

Anyway this is just food for thought. Feel free to post your ideas and suggestions to the mailing list, we will have an online meeting sometime in the new year - probably towards the end of January when my student's examinations are all over!

Thursday 9 December 2010

Better Late Than Never!


I am pleased to announce that Vinux 3.1 (Based on Ubuntu 10.10) is finally ready for download from the Vinux mirrors! It is currently available as a CD or DVD in both 32 and 64bit versions (USB and Virtual versions will follow shortly). It has been a long time coming, but hopefully it will be worth the wait. On top of all the usual accessibility software, new features include a Quick Start Guide for beginners (Ctrl+Alt+Q), Autokey-GTK which can insert text automatically as you type based on pre-defined abbreviations, the Parcellite Clipboard Manager which allows you paste text from the clipboard history, X-Tile which allows you to tile windows automatically, Gnome Media Player as an accessible front-end to VLC, Conkeror a keyboard controlled Web Browser, Pidgin the Internet Messenger (with all the plugins) and Gufw a simple but effective Firewall Manager.


Other additions include: Alarm Clock, a simple GUI for setting timed notifications, GtkHash an md5sum calculator, SearchMonkey an advanced search tool, Terminator which allows you to open multiple tiled terminals in one window, Tux Commander a dual-paned file manager with keyboard shortcuts, World Clock which allows you to monitor times in different locations around the world, Gcolor2 a simple GUI for selecting colours from anywhere on the screen, the Specimen Font Previewer, Dlume a simple address book manager, the ToDo List package, mhWaveEdit a simple sound recorder/editor GUI, Sound Juicer an audio CD ripper, FSlint a file system cleaner, Gtk-ChTheme a GTK theme previewer/changer, Scheduled Tasks a simple GUI front-end for cron and HardInfo a system profiler and system benchmarker.

There are also three new packages unique to Vinux available from the Vinux repository and/or the EasyInstall-Office and OCR scripts. These are: Markup-Binder written by Isaac Porat, which makes the creation of complex navigable documents a simple process (the new QuickStart Guide was created using this), another application is Speedy-OCR written by Don Marang, which is a simple GUI front end which allows you to scan in documents and convert then to text and/or speech. And last but not least is Monitor-Toggle written by Luke Yelavich which allows you to turn off your monitor to save battery power and maintain your privacy.

There are also some changes under the hood. We unfortunately had to remove YASR from this release as it was no longer working under Maverick, and in order to make room on the CD we also had to remove Samba (Windows File-sharing) and the Festival Speech Synthesiser, although these can easily be reinstalled after installation. The dynamic font-size/colour-theme changing scripts have been revamped so that the desktop background and desktop icons change colour as well, and we have gone with a dark theme by default, with the window controls in the traditional top-right corner position. In addition there are now both desktop icons and menu launchers to take users straight to the Vinux website and the support/development groups.

Now for the bad news. I have already found a small bug in the latest release - however it is unlikely to affect the majority of users and there is a very simple workaround for those who are affected. Because of a minor typing error in one of the build scripts (I used a lowercase 'u' instead of an uppercase 'U' - I was tired!) the keybinding (Shift+Windows+Alt+Up) which was supposed to modify the multimedia keybindings in the event of the sound card being muted wasn't enabled. The workaround is simply to press Alt+F2, type 'volume_keys' (without the speech marks) then press Enter, then press Alt+Y and Enter again. Now you can use Windows+Alt+Right to unmute and raise the system volume. Of course, I also managed to provide completely the wrong keybindings for the new multimedia volume controls in the keybindings.txt file as well, just to confuse matters (I did mention I was tired didn't I?). So to clarify, once you have activated the new multimedia keybindings (should you need to) you should use Windows+Alt+Right/Left to raise/lower the volume and Windows+Alt+Down to mute/unmute the volume. My OCD Demon is now muttering away about the fact that I should remaster Vinux 3.1 to eliminate this stupid error, otherwise the world will probably end. On the other shoulder my Sanity Angel is reminding me that I really need a good rest and if this is the only bug it has I should count my blessings and rejoice!

To be continued...

Monday 31 May 2010

Vinux: The Third Generation!


On behalf of the whole Vinux community I am happy to announce the 3rd release of Vinux - Linux for the Visually Impaired, based on Ubuntu 10.04 - Lucid Lynx. This version of Vinux provides three screen-readers, two full-screen magnifiers, dynamic font-size/colour-theme changing as well as support for USB Braille displays. Vinux is now available both as an installable live CD and as a .deb package which will automatically convert an existing installation of Ubuntu Lucid into an accessible Vinux system! In addition, we now have our own Vinux package repository (from which you can install our customised packages with apt-get/synaptic) and a dedicated Vinux IRC channel. In the very near future we will also be launching a Vinux Wiki and releasing special DVD, USB and Virtual Editions of Vinux 3.0!

Sunday 4 April 2010

The Day After Tomorrow!


Unlike in the ecological disaster movie of the same title, the future for
Vinux is looking very good! We are now approaching the release of the next version of Vinux, the third major release and there have been a whole host of bug fixes and new features included. The most significant change this time around is that the development is being led by Bill Cox instead of myself. This has been a positive move for two reasons: A. Bill's skills and knowledge of Bash Scripting, Gnome Accessibility and the inner workings of the Linux operating system far exceeds mine (I have never pretended to be anything other than an enthusiastic and dedicated amateur, who makes progress through trial and error - but mainly error!). And B. This means I have been able to sit on my fat arse and watch NCIS sometimes in the evenings instead of spending every waking moment hacking!

Bill has now sorted out most of the bugs that prevented us from using
Ubuntu as the basis for Vinux in the 2nd series: i.e. the lack of responsiveness and stability when using Orca, and most of the problems which were caused when using packages which required root permissions. Bill has modified the PulseAudio sound server to run in system mode and channeled all of the speech software; Orca, Speakup and Yasr to use Speech-Dispatcher and Espeak, which provides a seamless user experience when shifting between the three screen-readers. He has also reintroduced Compiz magnification in addition to the Orca magnifier which was left out of the second series based on Debian. Apart from the improvements to the accessibility of Vinux, there are also some new features such as EasyInstall scripts - icons on the desktop and in the menus which allow a user to install suites of packages such as Open Office or the Non-Free Multimedia Codecs with one click! We are also toying with then idea of including the Pcmanfm file manager which offers a much more responsive performance than the default Nautilus file manager. This edition will also see the reintroduction of the core CLI applications and the customised Bash aliases that were introduced in the Vinux 2.1 CLI editions. Of course the main advantage of switching back to Ubuntu is the improved hardware compatibility it offers especially for wireless connectivity which was always a problem with the Debian based 2nd series. While Bill has done most of the hard work other members of the mailing list have of course helped and supported Bill by submitting scripts or ideas, testing the alpha and beta editions or just providing moral support. So thanks to everyone who has contributed, I am not going to mention individuals in case I miss someone out.

Apart from the forthcoming release itself there have also been other significant developments. We have now been offered several more mirrors to host
Vinux Iso images on, and we plan to launch a new Vinux Wiki, Forum and IRC Chat channel alongside the release of Vinux 3.0 - so a special thanks to everyone involved in this process, taking on these responsibilities allows Bill and myself to concentrate on the development process. Perhaps the most significant development is the creation of a more mature Vinux Build Script and a Vinux package repository on Launchpad which allows anyone to turn a standard install of Ubuntu Lucid into a Vinux-like system. David Knight created the first Vinux build script, and this simplified the process somewhat, I then mutilated David's scripts to create the first CLI editions of Vinux, but Bill has taken this to a whole new level. Instead of being a single script with a few data files, Bill has created four separate scripts, three of which are used to turn an installed system into Vinux, the fourth to remaster the installed system and create a new Iso image. This means that nearly anyone can have a fully accessible system, without having to download and install Vinux, as long as they start with a default Ubuntu Lucid installation. The scripts are now hosted on SourceForge using Git so that other developers can get involved in the process. I managed to get the hang of it in a few hours (with a little coaching from Bill) so it can't be that hard - but you do have to know what you are doing if you don't want to destroy all of Bill's hard work!

Ubuntu will be releasing the 2nd Beta version of Lucid on Thursday and a Release Candidate on the 22nd April, The final release of Ubuntu 10.04 is planned for the 29th April and unless there are any unforeseen catastrophic bugs Vinux 3.0 should be released in the 1st week of May.

Tuesday 16 February 2010

The Future's Bright - The Future's Orange!


Over the last few weeks Bill Cox has made significant progress on fixing the problems caused by PulseAudio in Ubuntu. Some of these fixes have actually been ported back into the development of Ubuntu and at the moment it looks like this will provide a very responsive and stable accessibility experience. Although it is not quite as snappy as Orca with Alsa on Debian, it is very close and Ubuntu of course offers much better hardware compatibility. Coupled with the fact that Remastersys doesn't yet support Squeeze and it is very likely that the final release of Squeeze may be delayed even further, it looks like we will initially be going with an Ubuntu-only version for Vinux 3.0! This will also provide a simpler installation process, a USB installer and WUBI support. I have spent the last few days working out how to built up the iso from scratch starting with the mini.iso, so that it contains all of the necessary files but still fits on a CD. If there are no nasty surprises ahead then this will become the first LTS version of Vinux with three years support and updates. Bill has also created a Vinux repository so we will be able to provide patches, updates and Vinux specific packages through synaptic and/or apt-get. There will also hopefully be a DVD sized 'Leviathan' edition which will contain lots of extra packages, but it may not be possible to provide a pure CLI version as the Ubuntu version of Remastersys requires a GUI to work. The main release will however come with speakup and a basic range of CLI applications pre-installed. So I would like to thank Bill for all of the hard work he has done, as well as everyone else who has contributed to the development process by submitting fixes, suggestions and/or feedback.

Sunday 10 January 2010

New Year's Resolution: 2010x365!


So what does the new year hold for Vinux? Well I have just started work on version 2.0 of the Vinux build script which will allow the user to create a CLI, GUI or HYB(rid) version of Vinux from a standard Debian installation. It will allow you to choose between nine pre-configured builds or create your own unique customised version. The pre-configured builds include CLI, GUI and HYB versions of the 'dna', 'std' and 'max' editions. The main release of Vinux will be based on the HYB 'std' version providing a basic range of both CLI and GUI applications. Obviously the 'max' edition of the GUI and HYB versions will be DVD sized images, the HYB version containing all of the required, recommended and additional packages in the Vinux suite. The next release will be Vinux 2.2, which will probably be the last version of Vinux based on Debian Lenny (5.03). Once Vinux 2.2 is released and any bugs fixed I plan to start working on the Debian Squeeze version (Vinux 3.0) using the same scripts which will provide better hardware support and new versions of Gnome, Orca and Open Office etc. There is a lot of work to do and I estimate it will take me at least a month to get the script into a state were I can start making test builds - I hope to have Vinux 2.2 out by Easter if possible.

However that is only part of the story. There are of course other members of the Vinux Development Group who are working on their own sub-projects. Bill Cox is now working on a version of Vinux based on Ubuntu 10.04 Lucid Lynx. He is trying to resolve the latency and stability issues caused by the PulseAudio sound system and seems to be making very good progress. This means that there will hopefully be two versions of Vinux 3.0 - a Debian based version and an Ubuntu based version which will provide users with a choice between stability and cutting edge packages/features. Mobeen will keep producing the VMWare virtual editions of Vinux, and is considering setting up his own website/hosting platform and a Vinux User Forum to differentiate between the needs of users and developers. David Ring will hopefully be producing USB versions of Vinux and possibly a special CLI version based on INX. Don Raikes is working on some scripts with dialog front-ends to ease the process of partitioning, formatting and imaging hard drives, as well as a Vinux installer and possibly a specialised forensics version of Vinux. I am also hoping that David Knight will be able to contribute to the Vinux build script again once his work and family responsibilities allow him to. I hope I haven't missed anything or anyone out!