So I was working at the greatest place on earth (GPE) the other day, which of course meant the glorious opportunity to work with Greg the Redneck. He’s difficult to describe, but essentially enjoys tractors, fixing trucks and seeing stuff blow up.
So Greg the Redneck is a carryout, because the GPE is a grocery store. But on this day, Greg was hungry and bought an entire bag of fun size Twix and ate it in a span of twenty minutes.

Me: That’s bad, Greg. Twix is the most unhealthy candy bar you can eat, and you just ate 15 times the recommended amount.
Greg: Baaaaah I’m HUNGRY.
Me: Settle down, Greg.
Greg: I’ll TELL YOU when I needa settle DOWN!

Arguing with Greg is a losing battle. So instead of lecturing him about eating healthy, I got him to agree to star in a made for T.V Christmas special called “The Ghosts of Diabetes Past.” Or possibly, “A Very Greg Christmas.” Here’s the premise:

Greg is a down on his luck redneck who works at a grocery store. He never feels great because he has an awful diet consisting of twix bars, marshmallows, root beer and squirrel. After devouring a bag of candied possum for Christmas Eve dinner, Greg falls asleep. Immediately he is visited by an auspicious ghost.

This ghost is the ghost of diabetes past. He shows Greg all the terrible mistakes he has made with his diet. He presents Greg with a vision of when he was six and he drank a bottle of corn syrup.

Terribly confused, Greg is then visited by a second ghost. This ghost is the ghost of diabetes present. He shows Greg what his pancreas looks like; it is withered and purple and intermittently on fire. Greg is stunned by his poor health.

The third ghost, of course, is the ghost of diabetes yet to come. He shows Greg when he is 38 years old, morbidly obese, and missing a foot.

And thus, Greg the Redneck learns the true meaning of Christmas: Don’t eat too much candy or you’ll lose your foot.

Its time to make announcements… I should have done this earlier, but like always it takes me a while to find a free minute to write things down.

First is about OpenEmbedded: We will be present at LinuxTag from 24th to 27th of June in Berlin/Germany. I just hope they have a little bit more space for us at the Embedded Area compared to last year. We usually have a lot of interesting devices running OE built Linux we can show. I’m really looking forward to LinuxTag – I haven’t missed one for many years and its always a nice event with a good mixture of community members, business people and users.

LinuxTag

Google Summer of Code is in progress for some weeks now and I’m in the great position to mentor a very interesting Maemo project… something I always wanted to see but noone found time to do it properly so far: Integrate the Maemo software framework into OpenEmbedded. The benefits are quite obvious – Maemo software will gain compatibility and quality by running on (and if necessary adapting to) various devices. Apart from the fact that Maemo is a pretty good open source framework attracting quite a lot of developers. This is something other devices can and should gain advantage from. Well rkirti made a pretty good description of the project which can be found here.

It would be pretty cool to see Maemo running on this nice device I received a few days ago…

Micro2440 running gpe-mini-browser2

Its a FriendlyARM Micro2440 from Watterott Electronic – this one is equipped with the 7 inch display which gives some more “freedom” to the applications Apart from this the hardware is the same I described in my previous post.

The screen geometry would match the one used by all Maemo devices so far, so it would not mess up all the graphical user interfaces. But there is still some work to do till the boards become a really good development platform. We can build useful filesystem images for them but the installation is still split up into too many steps and the up to date kernels still lack proper support for the camera and the wifi module.

I have made a toolchain to build software for both these devices and the Topas910. Together with an updated install instruction it can be found here at labs.kernelconcepts.de.  Now I’m investigating ways to integrate the toolchains with IDEs in order to simplify application development for mobile and embedded devices.

Micro2440 with Ultimate++ demo application

Here the device runs one of the Ultimate++ demo applications built with its IDE and the GPE-flavour cross toolchain built with OpenEmbedded. A friendly colleague found out what needs to be done to use it for ARM targets… it still needs a little help since Ultimate++ doesn’t seem to have an idea about pkgconfig and I’m not really happy with the size of the resulting binaries. But more about IDEs later…

Have a good time… and see you in Berlin!

While we are at the topic ‘evaluation and development hardware’… here here a few lines about a similar piece of hardware. I promised to do something with a FriendlyARM board already – so I did. I choosed the Micro2440 because the board itself is very small and can be used for own design easily. The very similar Mini2440 has a lot of ports on board in contrast to the 2mm headers of the Micro2440. Both boards are available through local distributors in Europe and the US now so that its not necessary to place orders in China any more. Watterott Electronic was so kind to send me a Micro2440 board and the SDK baseboard for evaluation.

If you want to get started with embedded Linux the board might be exactly what you want: A 400MHz ARM9 based Samsung S3C2440A CPU which is pretty well supported by Linux. 64MB od RAM and the same amount of NAND flash and 2MB of NOR flash for a backup bootloader. The baseboard has the connectors for the serial ports, USB host and client, display, expansion and SD and adds an audio codec and an Ethernet chip. There are two variants available: One with a 3.5″ QVGA TFT (pictured) and one with a 7″ 800×480 TFT.  Both displays come with a touchscreen and in addition to this the board has a few buttons that can be used for a human interface. The whole SDK kit including Micro2440, SDK board, 3.5″ display cables and JTAG doesn’t cost more than 125 EUR incl. VAT which makes it even more appealing.

The ’softer’ part of the SDK quite appealing too: The hardware is pretty well documented, even the schematics are available to the public. FriendlyARM released some demo images and Linux sources that are useful (but not perfect). The ‘Vivi’ bootloader used by the boards can be replaced with u-boot easily so that you get a 100% Open Source embedded development environment.

Thanks to the OpenMoko community the S3C is pretty well supported in Linux and u-boot. There is a public project providing up to date Linux, u-boot and QEmu supoort for these boards that can be found here. Like I usually do for a new piece of embedded or mobile hardware I used OpenEmbedded to built a GPE based filesystem image for it. Like you can see in the picture it works pretty well. For people interested in this I put together some notes I took and the binary images at KC Labs. Feel free to contact me about this… I plan to extend support for it a little bit and provide a toolchain and additional information.

Have a nice time!

PS: Sorry for the bad image this time, I had to use the webcam to capture it.

Some time ago I got a nice looking and interesting TOPAS910 evaluation board for the ARM based Toshiba TMPA910 series of SoCs. With its black PCB and golden labels for the expansion ports it is definitely the best looking evaluation board I have seen so far. But its interesting from a technical point of view as well and of course you can run Linux on it.

Topas 910 Board

The TMPA910CRAXBG SoC on the board is based of an ARM9 core clocked with 200MHz and a pile of external interfaces such as:

• TFT controller
• USB client
• NAND and NOR flash
• SPI
• Touchscreen
• ADC
• SD / MMC
• UART

The board can be powered from the USB port, has Ethernet, a QVGA TFT display and all necessary interfaces to get started with the SoC.  Some details can be found here.

Expansion Connectors

The initial Linux port was made by a company called BPlan known for their Amiga projects. They provide two ports: A real port to the hardware platform and one making use of some proprietary OpenFirmware-like bootloader. The latter is quite useless because it depends on calling its firmware for every hardware access but the 2.6.26 patch, bootloader and documentation they provide on their website are a good start. After a deeper look into the kernel patch it was clear that they left a lot of things to do: Only half of the TFT, NOR flash, USB, Ethernet anf the serial port are supported. It among other lacks a driver for SD, NAND, audio, the joypad and all the GPIO ports.

I wasn’t able to resist and started a small project to improve support for it. I have updated the kernel to 2.6.29, added gpiolib support and drivers for joypad and the small LED display which helped testing the GPIO drivers. Of course I have OpenEmbedded support for it ready and pushed upstream today – even if this still uses the old kernel it is useful to build a tiny userland that is able to live in NOR flash. Now I need to get some storage facility sich as SD or NAND flash working in order to deploy a proper filesystem image.

The TMPA910 is the BGA case in the center of the image

I would be interested in getting to know any hardware that is using this TMPA910. The results of my work: Sources and a small demo image can be found at KC Labs. We haven’t announced it yet, but its there already: KC Labs is the new Open Source projects website at kernel concepts.

Enjoy!

Some time ago I wrote some lines about the nice combination of projected human interfaces such like the TI PICO projector for visual output and the ‘laser keyboard’ for input. The innovation you can gain from such a combination is a device whose interfaces dimensions can can differ from the dimension of the device quite a lot. After playing around and having an interesting time watching people using these technologies (e.g. at FOSDEM) I finally manged to create a device study combining the components into a single case. It turned out to be more complicated than I thought and the way to a ‘production ready’ device would involve a lot of research and improvement.The main idea was to have something you can touch and try if it can be useful.

But let’s take a look at it first – for now its just a kind of brick with a BeagleBoard and some more components inside.

"The Device" in action

From a technical point of view the contruction is quite simple – its just a collection of easy to get components:

• TI BeagleBoard (Rev. B4 in this case)
• Celluon CL850 “Laser Keyboard”
• A tiny USB hub
• TI PICO projector
• Some custom USB cables
• Huawei UMTS modem
• One Bopla BOS 800 case
• A small RS-MMC card for the root filesystem

The batteries didn’t fit into the case – the same problem like my GSM evaluation platform I intended to use for connectivity. Instead of this I had to use the Huawei modem but lost an easy to use audio part (I still would like to see these interfaces in some kind of smartphone) and the charger for Li-Ion batteries. The USB connected UMTS modem is advertized to have audio capabilities but there is no Linux support for this feature yet. Another component that caused some headaches is the HDMI cable for the PICO projector: It is thick and its big plugs waste a lot of space in the small case.

I removed the cases of USB hub and keyboard to make them fit into the case. From the BeagleBoard I had to remove the S-Video connector in order to reduce its height.

The current software is way less spectacular than you might think. Its a simple Ångström distribution GPE image built with OpenEmbedded. I had to replace Xorg with Kdrive in order to get the Xrandr extensions working and added some more software for testing such as an additional browser. This is not really the software you wold expect for a modern smart phone but its a good environment to test an uncommon mobile computer.

So what did I find out in the initial tests?

• It looks very very geek!
• Its useful to some extend but you need an even surface.
• The mechanical design needs improvement. This ‘brick’ case is not really flexible and even a better arrangement of components (e.g. projector and keyboard side by side) would save a lot of space on the table.
• It would be necessary to have a different lense for the projector: You want a wide angle lens to get a large projection area in a short distance drom the device. As you can see in the image the desk space you need is quite high but the image is still small.
• The power consumption of the projector is a major problem. It is hard to power with batteries and in a plastic case it gets very hot.
• Someone needs to come up with a clever holder for a sheet of paper to make it a useful screen. Suggestions welcome!
• If I have the chance to do so I would like to try an improved prototype with a small secondary display, batteries and a keypad.

Typing on the table

A device I could imagine would be a kind of a stand for a smartphone with built-in projector module. The next generation of projector modules should be small enough for this and comsule less power. The stand would allow the phone to stand upright on the table and provide the keyboard functionality. The advantage of this solution would be that you would not have to carry around the keyboard engine all the time. And because of the orientation of the phone you can get larger ’screens’ if the projector is in the top end of the phones case. Maybe I should sketch this for the next blog entry…

Have a nice time!

Coming back from my summer holiday (all of one week) I discovered Neuren Pharmaceutical’s share price had plummeted from 5 to 1 cent. Why? Its lead drug candidate, Glypromate/ GPE recently had failed its Phase 3 trial. Not a great way to start the New Year. Neuren released the announcement on Christmas Eve – quite sneaky considering the majority of its shareholders were probably down on the Coromandel cut off from the internet. It stated that their pivotal Glypromate phase 3 trial for the treatment of cognitive impairment in patients undergoing cardiac surgery with bypass was unsuccessful, that is, GPE exhibited no significant difference between treated and untreated individuals. Strangely only 20% of the 325 patients exhibited cognitive decline after surgery (i.e. they were only testing 64 people total with placebo or drug). They mentioned this number (20%) was a much lower rate previously reported in the literature. A quick literature search only brings up a handful of papers relating to this cognitive decline, most of it coming within the last few years. The makes me wonder whether this was the best way to actually test the drug’s neuroprotective capability. Part of the pull of this condition was that there were no current treatments available enabling the fast track to Phase 3 and hence saving $$. However the trade off was that by bypassing Phase 2b this meant the drug’s efficacy in humans was still unknown for a condition that lacked a well established scientific history/background.

Neuren is or was seen as NZ’s leading biotech/pharma company since the downscale of Genesis R&D, and was hopefully going to be NZ’s first biotech success pioneering the way for other life science companies. Where does the pioneering burden lie now? LCT? CODA? PROACTA? Pathway Therapeutics? More so where to from here for Neuren? Unsurprisingly GPE’s development will be discontinued. However they do intend to develop the rest of Neuren’s pipeline. The underlying question now becomes – are they able to raise the funds to do so while carrying a dented reputation. This combined with the ‘crunch’ puts Neuren in a very difficult position, with investors possibly seeking less risk and organisations with a proven history. For all we know this positive stance is simply to convey a sense of confidence in their strong pipeline for the means of attracting a buyer. We have all heard about the potential shopping spree by big pharma under the current economic climate. And at one point NZ$1.3 million (now up to 2million) could buy you a NZ biotech company with 6 pipeline drugs.

The Pipeline

Motiva – acquired last year as part of the Hamilton Pharmaceuticals buy out. “A novel cyclic (GABA) derivative for neuropsychiatric and neurocognitive symptoms of stroke and other acute and chronic neurological disorders including Parkinson’s disease and Alzheimer’s disease.”- Ph2b

NNZ -2566 – neuroprotection – Ph2b

NNZ-2591 – learning and memory for dementia – Ph2

NNZ-4945 treats neuropathy – Preclinical (PC)

NNZ-3006 – Obesity PC

NNZ-8000- anti-TFF for cancer PC

The Share Price Fall – From 5 to 1 cent.

Related Links

The Neuren story http://www.sharechat.co.nz/features/unlimited/article.php/8e17929b

____________________________

Graeme

– graemefielder.com -

]]> http://fl0rian.wordpress.com/2009/01/19/fun-with-beagleboard-and-projector-devices/ Mon, 19 Jan 2009 00:34:42 +0000 Florian http://fl0rian.wordpress.com/2009/01/19/fun-with-beagleboard-and-projector-devices/

A PICO projector is a nice thing: An interesting toy, useful tool or whatever you might call it. I got one of these from Textas Instruments in order to prepare small demonstration of an idea… but more about this in a later article. The most intersting question about these PICO projector engines is what they mean for the design of future devices.

From a mechanical point of view they offer a very important improvement: It offers display sizes that are not limited to physical size of a device. This is an interesting perspective for mobile devices. Will we finally get smartphones that provide a display which is useful for office work and internet browsing and still small enough to fit in a pocket. (I have to admit that I carry a quite modern/big smartphone – a Nokia E90 – in my pocket: It fits, but I have to say that there is some room for improvement.

When I read the announcement of the PICO projector at the BeagleBoard mailinglist another device I read about some time ago came to my mind: A projection keyboard.

Keyboard projection on desk

Even if it takes a little bit of time to get used to the projection keyboard it is an interesting alternative for mobile use. If its well calibrated for the surface it is a quite usable replacement for a real keyboard and still fits in a pocket. But more important for me: After a closer look at the construction I am convinced that it would be possible to reduce the size quite a lot. You see, I have mobile devices in mind all the time

BeagleBoard with Projector and Keyboard

The BeagleBoard is a nice platform because its performance and interfaces are similar to what we will have in quite some mobile devices soon. Its based on an OMAP3 SoC which is designed for powerful mobile devices and offers all sorts of peripheral controllers like USB host and OTG, TFT controller, SD interface and SPI. Several companies announced OMAP3 based devices already.

For my initial experiments I used the devices you can see in the image above. Apart from the BeagleBoard, a small USB HUB, the projection keyboard and the PICO projector there is a cheap USB Wifi adapter, a serial cable and a sheet of paper. Think about adding a powerful battery (power consumption is a very important topic, especially regarding the projector and the keyboard) – you would end up in a quite small device already. Not too bad, compared to the popular Netbooks but still with some drawbacks and a good mechanical design might be a challenge.

The BeagleBoard here runs a custom Linux filesystem with some GPE and other GTK+ based software built with OpenEmbedded. The latest 2.6.28 kernel OE builds seems to work pretty good, one major improvement seems to be the reliability of the USB OTG port which caused a lot of trouble in my previous tests.

Now I’ll add additional componets to get a more complex environment for playing with components and software I assume to be relevant in near future…

I recently spoke with one of my old teachers at Western Piedmont Community College about the subject of roaming profiles, which would allow the staff at cogcon use any computer at the office and have access to their documents.

I eventually didn’t do “roaming profiles,” per se, but just Folder Redirection.

He showed me how to get to the Group Policy Editor for the domain, and then showed me around, explained some of the useful settings.

Later, I went to cogcon and explored the GPE and found all kinds of cool settings and options.

I went to the Active Directory Users & Computers area, added a test user group, and added a few test users to that group.  I then set the “My Documents” folder redirection to go to
\\cogconsvr2\profiles\%username%\My Documents
After that was done, I tested it, and on many computers,  it still showed the local profile was in effect, because I had already logged into those computers before with those user names.  I tried restarting them, and it still did not work.  I then discovered a setting that told the computer to wait for the network, and after restarting, it worked.

Then, I went back to the Active Directory Users & Computers, added two new user groups, one for each office, and then added the respective users to their groups.

Then, I went to the GPE and set “My Documents” folder redirection for each:
\\cogconserver\profiles\%username%\My Documents and
\\cogconsvr2\profiles\%username%\My Documents
 Two computers in each office (four total) oddly enough would not accept the changes, and insisted on keeping its original setting for the documents, so I just set them manually.

I am not sure of what is special about those four computers, but I did not worry a whole lot about it.

During this project, I renamed all those “Workstationnn” folders to their respective owners’ names, which made things look much better.

I also created a script to map several drives that most of the staff will find useful or necessary, including the map to the AltaPoint data, QuickBooks data, etc.

I spend quite some time with Linux on all sorts of interesting devices. There are several ones not yet covered by my blog, so I decided to start a series of postings about devices running Linux. I introduce a new ‘Devices’ category in my blog to cover these hardware related topics.

Currently I am building a GPE filesystem image for the BeagleBoard I won at LinuxTag. They just announced that the board is on sale now.

For the friends of old Geode based web terminals such as the IBM Netvista J320/8365 I have an updated Angstrom based CF card image. It fits on a small 128MB card and comes with GPE and several browsers (gpe-mini-browser, midori and netsurf) to play with. In addition to this external USB touchscreens are supported. The CF image can be found here – if you want to build it yourself you can get kernel and filesystem seperately.

Another new ‘victim’ is the Neon board by Boundary Devices…. but more about these later.

Neon Board

Good news for me… after missing GUADEC and other interesting events it looks like I’ll make it to the first Maemo Summit in Berlin. The list of participants is quite impressive – I guess this will be a really interesting event. Just join us there

I read a few lines about odeviced… anyone else who thinks that using something like this for Maemo might be a goo idea?

I do not have much time left for blogging and coding currently – family and work keep me busy these times. But a few good things are in prgress – OpenSync’s roadmap indicates that they are close to a new release. This will be a much better base for MaemoSync than current SVN trunk. Even GPE makes a little bit of progress. Graham continues fixing various PIM bugs and gpe-memo is close to become ready for its first release.

Have a nice time…

Linke every year we had a really good time at LinuxTag! It is always a great event talking to interested users, business people and other developers. For the projects sharing this booth such as MyStep, Jalimo, GPE and OpenEmbedded it was a great success. For me it was the first time representing OpenEmbedded at a fair like this – with good results: We had quite some people who knew OE and asked specific questions and many developers with interest in using it.

LinuxTag always gives a lot of inspiration for new things to do and projects to check out. I know about some interesting software we need in OE soon – things like the Sugar framework for the OLPC and x2go. Other important tasks are to publish a list of evices that are defined in OE metadata and pushing the mobile-linux project forward.

In the Maemo track I had a short talk about GPE applications for Maemo – the slides of my talk can be found here. Unluckily I was not able to attend more of the interesting talks in this track. But I found out that the Mamona people seem to like OpenEmbedded at least

The Tarent and Mobile booth right after opening in the morning… it was way more crowded a short time later.

Many thanks to all the people organizing LinuxTag and Tarent for booth and the big red sofa for sitting down to write my blog.

Purchased a new “toy”.  The Nokia n800.  After thoroughly researching the product, I couldn’t help myself.  Apparently, this still was more of a compulsive buy than I’m willing to admit.  Got the device and quickly found that it was lacking much of what I envisioned.  First off, I tried using it out of the box with OS2007 and the email client (either though it claims to support IMAP) would not do IMAP and POP email was very, very poor.  I knew it didn’t come with any PIM-like apps (calendar, to-do, etc) but I was willing to work with what I could find on the net.  After putting the thing away for a day, and coming back to it, I started to make some great strides.  I updated the OS to OS2008 and started to Google links with n800 apps and other items like calendar, email, etc..  Bottom line is I’m very happy with the device now.  Claw-mail is a nice email client, and GPE development has come up with some straight-forward PIM-like apps that play nicely with my Google apps ..email, calendar, etc.  Anyway I will likely have much more to say about the n800, but as it stands today, it rocks. 

I have made an installable package for Maemo Sync, but do not expect too much. Its a basic port of the current Multisync-gui to Maemo. It is only tested to so far that the GUI starts up and registeres correctly. The fact that I had to use cmake for building Maemo software caused some headaches here… the quality of the source distribution package is still quite bad.

I also did not decide if I want it to become a project forked from Multisync-gui or to maintained with Multisync-gui adding optional support for Maemo environment. The always changing OpenSync API would be a reason to stay with Multisync-gui, but the fact that I have different opinions about UI design, Glade is involved and cmake are reasons for making it a separate project.

Binary package and sources are located here. You also need the OpenSync packages from Graham’s daily repository.

Another minor improvement of the GPE application packages in this repository is that latest Starling supports OGG playback. You only need to have an OGG plugin for GStreamer installed. The mogg package provides this for example.

Any feedback is welcome – if you manage to sync data with either the command line tool msynctool or using Maemo Sync please drop me a line. I would like to collect information which sync peers work with latest OpenSync and how to set up these.

Enjoy!

LinuxTag 2008 will be great for people interested in mobile and embedded Linux. I do not know about all the projects that will have a booth there, but GPE, GPE Phone Edition, QuantumSTEP and OpenEmbedded will be there at least. The application for a booth for these projects was accepted some days ago.

LinuxTag is one of the largest and most interesting Open Source events in Europe. It is known to be a great forum for meeting all sorts of people interested in Open Source – LinuxTag attracts developers, business people and users in equal measure. It will take place in Berlin from May 28th till May 31st

A big thanks to the LinuxTag people and the booth sponsors for the opportunity to show what we are doing.

For OpenEmbedded event planning we have a wiki page that can be found here.

Some weeks ago the USPTO rejected the GPE trademark again – finally!

This is great news for the GPE project which really suffered from this situation. The latest entries in the ‘prosecution history’ look nice:

2008-02-27 – Notification Of Final Refusal Emailed
2008-02-27 – Final refusal e-mailed
2008-02-27 – Final Refusal Written

The complete attempt by Mr. France to register GPE as his personal trademark is publicly available at the USPTO website. Some background information can be found in this article. The handhelds.org website still claims several trademarks, but the related project become more and more obsolete. The hh.o bootloader is used for ancient hardware only or replaced by U-Boot and ipkg gets replaced by the much better maintained OPKG. Its a little bit sad that the Opie projects are not that active any more though…

I made some progress porting the multisync-gui to Maemo. The idea is to offer an easy to use graphical user interface for synchronizing data between the Maemo based tablets and other devices. The OpenSync framework that is used offers several plugins for interfacing data sources already. The most interesting ones are the two SyncML ones (HTTP server and client) and the GPE plugin we can use for the PIM applications on the Maemo device.

Currently it is far from beeing useful, but there are some windows to make screenshots now The first thing I want to archieve is to be able to sync PIM data from a Series60 cellphone.

Speaking about Series60 – I received my Forum Nokia Champion diploma today which came with a great target device for syncing experiments Many thanks to Nokia for this nice reward and for the great Forum Nokia Champion program and to Nils who suggested me for it!

Did I ever complain about the fact that days are too short to write about interesting things? Time to do so… but well, finally I managed to write a short summary about latest updates:

Thanks to Nora we have a major improvement of the design of the GPE website. I even managed to update the PDA image on the index page last week. GPE is going to work on Maemo Chinook as soon as it is ready I guess – Graham and Christoph are doing a good job sorting out the remaining issues and building new binary packages. I’ll work on this a little bit soon I guess, I’ll get a N810 from the Maemo device program as well.

The new kernel concepts website is online now and finally has both German and English language support as well as a much better look&feel and a nice new shop. The only major task remaining is the English translation of all the shop contents.

There are some more updates… family updates, workstation upgrade (Intellistation M-Pro to Z-Pro) and a lot of minor updates, but nothing that is really relevant to the public

Have a good time…

I got a nice loveletter from Mr. France of handhelds.org today – I do not think that I should quote him here But its a good reason to point interested people to two links. The first one is the letter of the USPTO in response to the GPE trademark application from 25. of June. A short summary of that letter is that hey did not accept the screenshot of the old hh.o GPE website as proof for GPE beeing used in commerce by hh.o and request a better specimen for a “GPE product”.

The second link is the page describing handhelds.org itself. The section “What we do not provide” gives us some information that hh.o does not sell any products – no hardware, no software and they do not even provide support. (In the case this page has some “accident” or suddenly changes I have a nice screenshot I can provide.)

So what’s the point here? Is there a secret hh.o product that suddenly pops up to fulfill the requirements for a trademark?

There are more reasons why this whole trademark story smells fishy:

• The date for the first use in commerce stated in the trademark application is a long time before hh.o Inc. was founded, at a time the platform was part of the CRL and the domain belonged to Kelly Martin. So who did this business? Compaq?
• Even if the hh.o website tells us that hh.o owns the GPE trademark this is not entirely true: The trademark was filed by George France personally and not by the non-profit handhelds.org Inc.
• Mr. France was never involved in GPE itself, only in administrating the servers used by GPE. Noone ever “donated” GPE to anyone and he simply has no right to claim GPE for himself.

Someone came up with the argument GPE was forked and did never move to LinuxToGo when this dispute started… well my answer was that future will give us clue about which of these opinions is right. From the time we moved till I was unsubscribed from the old GPE commits mailinglist today there was a single mail on this list – caused by Mr. France himself removing a file from the GPE CVS repository. I do not think that any honest person is able to speak about a fork in this case…