I just announced the death of idjvu over in the idjvu thread.

My plans for this weekend are to refactor ipdf into a pluggable viewer framework.

My plan is to release 3 viewers with the new ipdf.

1. wrapper for Poppler e.g PDF support
2. wrapper for DJVULibre e.g. DJVU support
3. wrapper for my RTF parser e.g. Baen RTF ebook support (and yes, I'll still be doing the read direct from ZIP trick.)

The ipdf will use direct to screen memory page transfers with an X window popup place holder for it (so X doesn't get confused.) It will have a full 8 bit path for crisp text.

Support for improved decoding time and error dispersion of images will be in the next release of the Poppler wrapper (sorry, only so many hours this weekend.)

Good luck with yor project scotty.

I've been following your progress with keen interest!

I (and no doubt many on this forum) apprecaite all the time and effort you are putting into this. I'm sure something useful will come out of it! ;)

I just announced the death of idjvu over in the idjvu thread.

My plans for this weekend are to refactor ipdf into a pluggable viewer framework.

So you will ease the creation of new viewers for the rest of us. Thanks :-).

The ipdf will use direct to screen memory page transfers with an X window popup place holder for it (so X doesn't get confused.) It will have a full 8 bit path for crisp text.

Uh? What's that popup thingy?

Support for improved decoding time and error dispersion of images will be in the next release of the Poppler wrapper (sorry, only so many hours this weekend.)

Anyway, big thanks for your work :-b.

Sources?

Can you publsih then idjvu sources? While you are tweaking new pdf, we will be happy to user old djvu, but with zoom and without black band in the right.

I just announced the death of idjvu over in the idjvu thread.

My plans for this weekend are to refactor ipdf into a pluggable viewer framework.

My plan is to release 3 viewers with the new ipdf.

1. wrapper for Poppler e.g PDF support
2. wrapper for DJVULibre e.g. DJVU support
3. wrapper for my RTF parser e.g. Baen RTF ebook support (and yes, I'll still be doing the read direct from ZIP trick.)

The ipdf will use direct to screen memory page transfers with an X window popup place holder for it (so X doesn't get confused.) It will have a full 8 bit path for crisp text.

Support for improved decoding time and error dispersion of images will be in the next release of the Poppler wrapper (sorry, only so many hours this weekend.)

The cold hard realities of the ashes of hope are settling in.

Collaboration with iRex is going nowhere. I'm working on software that has a high potential of bricking the iLiad of someone that doesn't exercise enough caution with its installation. And iRex did not design things like er_registry.txt with 3rd party software installers in mind.

The pluggable ipdf is all good stuff from a technology view though.

I inspect ZIP files. A Baen RTF file gets paged into Pango (solved some issues for me from my previous backend PDFBox). A collection of JPG/PNG's gets paged as a graphic novel or you can put a slideshow time value into the manifest.xml and it flips pages automatically as a slideshow.The same self advance value can be applied to PDF's and DJVU's as well.

I've figured out where to modify Poppler to do the enhanced image ED and accelerate CITT decoding via libtiff. But swapping out libpoppler has a very high pucker factor.

Even my x48 port requires a dangerous edit to the er_registry.txt to install the SHELL hack.

With no means of safe software distribution its all beginning to look very pointless.

So, once again we need to wait for iRex to give us an "unbricking" option or, even better, a "legal" way of installing software. There was a post at their forums today indicating that they are qorking on that solution, but I wonder when we'll see it...


Here is the link to the iRex statement:

http://forum.irexnet.com/viewtopic.php?p=3166#3166

With no means of safe software distribution its all beginning to look very pointless.

Isn't that the nature of developing for an unsupported system/hack...? (Though its certainly interesting seeing what the Illiad is capable of...)

So, once again we need to wait for iRex to give us an "unbricking" option or, even better, a "legal" way of installing software. There was a post at their forums today indicating that they are qorking on that solution, but I wonder when we'll see it...


Here is the link to the iRex statement:

http://forum.irexnet.com/viewtopic.php?p=3166#3166
As per their usual: long on concept, short on details and totally devoid of delivery dates.

"they will learn patience."

:)

They must remember why they need us.

:)

I think it is a shame that the initial good-spirited communication and the incredible accomplishments (in such a short time) seem already forgotten.

I'm more than willing to help this community with their development (even though I'm incredibly busy working on the next release), but there are some things I cannot just 'magically' make appear or happen.
Even so, I think that if we keep communicating and work around the problems together, we get a lot further.

If there are any reasonable requests or questions related to this, please post them on the iRex forum and I/we will react, as I've done to date with all such requests that were posted there.

I think it is a shame that the initial good-spirited communication and the incredible accomplishments (in such a short time) seem already forgotten.

Exactly, the paraphrase above is taken from V for Vendetta as the dictator exhorts everyone to greater efforts to stop the "terrorists".

So far I've helped show how to speed up the PDF reader and deliver sharper text at the same time. I've developed viewers for three new file formats for the platform. Delivered a much needed screen capture program, ported a calculator emulator to the e-Ink panel and worked to deliver safer means for people to execute all the above. I've also been busy working up PXA-255 MMX primitives to speed up additional operations. All of which you were aware of.

What has iRex done since then, other than try to lock people out of using the applications developed? You know how many emails I'm getting from people unhappy with being on 2.7.1? Emails from developers whom paid to get their iLiad's re-flashed when their e-Ink panels failed (not re-flashed because they bricked their units via development) and got back 2.7.1?

People begging me to give them an exploit back into their iLiad's?

Have I been running around giving out exploits?

No, because I figured it would upset you all to have to keep diverting efforts to keep locking us back out of the box.

If I truly had negative feelings I'd be making your lives miserable by slowly releasing exploits to speed up your release cycles. :D

For my part I couldn't even get the Poppler version number out of you! I had to spend time guessing what base version to work from. And frankly the only reason for it was to spend countless hours polishing the graphics primitives to support your firm's business issues about the quality of embedded images in PDF's.

Cooperation is a two way street. What good is a SDK when only a crazy few can run the code it generates? And those crazy few are constantly walking a tight rope above a chipper shredder waiting to claim them and remove their ability to continue working on code.

Such an environment separates the weak from the strong, and the strong tend to not be shy about sharing their thoughts about the cloud of virtual shredded corpses swirling around them.

What do we want? We'd like a re-flash utility.

Turn off bricked iLiad: push reset button.
Hold connection button in upper right.
Slide power switch on, hold until special blink sequence on LED.
Release power switch, wait for iLiad to turn back off.

The re-flash would grab a factory clean user filesystem image off the /mnt/free in the M-System's DOC and re-install it into the user filesystem. Restoring the iLiad back to factory original and boot-able state.

That's a cheap 20MB safety blanket to leave in /mnt/free.

We'd like either a chroot -u sandbox -g sandbox [SHELL] enhancement to ContentLister to execute files with extensions of .sh or .SH. Or a chrooted Sandbox that the iLiad would run on startup out of /etc/rc5.d/S99sandbox.

You can examine the ZIP file I created and uploaded as an example of how gnuchess could be packaged to not need any root user, or root filesystem, access.

I'll get some sleep before reacting to this big post. But:
For my part I couldn't even get the Poppler version number out of you! I had to spend time guessing what base version to work from.
I am not aware of a request from you about the poppler lib version number or anything related, so I don't understand this remark at all. :blink:

Cooperation is a two way street. What good is a SDK when only a crazy few can run the code it generates? And those crazy few are constantly walking a tight rope above a chipper shredder waiting to claim them and remove their ability to continue working on code.


Which SDK is this...? :blink: So far I've only seen Irex release source & tools... I thought the SDK was still being finalised...?

Which SDK is this...? :blink: So far I've only seen Irex release source & tools... I thought the SDK was still being finalised...?

Jaed, the "tools" contain all (or most of) the libraries plus the headers, it is sort of a mistake that they announced it simply as a "chaintool" and thus some people has not looked into it. Of course, one still needs documentation and some more examples. And of course, a full SDK would be something more on the way of Scotty's "pluggable ipdf", with generic g++ classes etc.

Exactly, the paraphrase above is taken from V for Vendetta as the dictator exhorts everyone to greater efforts to stop the "terrorists".

I do not remember the quote. But it is a fine coincidence that the first software compilations have been released in the weekend of the 5th of November, year 401 from the Gunpowder Plot. Guess iLiaders will not forgot the day :)

Please, calm down everyone...

We have been through a lot together and most of it has been hard times, but let's not get into a fight here...that will help nobody.

The way I see it, iRex badly needs to get some of the development work "out of house" so that things will start to move faster in the completition of the coreOS and features.
Scotty has a big point here, since he's done a lot of hard work FOR FREE and managed to do a great deal working with little to no information and cooperation from iRex.
The trouble is, that he is woring for a very select few users because most of us (me included) can't actually run his great little programs and improvements because we have 2.7.1 and/or are afraid of editing dangerous files on the Iliad.

Now, to us it looks like iRex is working on closing the "holes" to prevent people from using the community-created software. Of course we understand that it is to keep us from bricking our Iliads, but what we don't understand is why we don't rather get a "legal" way to install and run those apps. The information required to create a reflash option or to make a safe install possible has to be available to iRex, but apparently you don't have the time to create a woring solution.
This is OK, we understand that you maybe have priorities elsewhere....
But why don't you at least furnish a few select developers with this information and let them help you? Why don't you give the full sources (which are mainly opensource anyway) and maybe some device-information neccessary for the work to a few select developers here? Let them sign a NDA, make the sources available to them alone for the time being or something...whatevery you do, just let those who want to help you actually help!
You website is full of promises about the "openness" of the platform, something that brought many of us round to buying the Iliad instead of the "closed" platforms like the Sony or the Hanlin V8. In reality, se have seen little of that open platform. We have seen promises of the SDK and sources, we have had dates for their release, but we haven't got anything but a few snippets of code even now. Every bit of information has to be pulled from iRexs' grasp with a lot of patience and effort (look at the battery-life/Suspend mode fiasco where people had been asking for months without a reply before you finally released a statement...and never followed up on the question that arose there...) and then you wonder why the user-community is getting hostile?

Jaed, the "tools" contain all (or most of) the libraries plus the headers, it is sort of a mistake that they announced it simply as a "chaintool" and thus some people has not looked into it. Of course, one still needs documentation and some more examples. And of course, a full SDK would be something more on the way of Scotty's "pluggable ipdf", with generic g++ classes etc.

But doesn't a SDK imply that there's somewhere to run the software created with it...? While I could write a Universal Reader E-reader for the Iliad its a bit academic until I can run it somewhere... And the source code was released after 2.7.1...

Not to mention that a full SDK would have the examples + docs (in an ideal world :) ) you mention.

But of course, we could be back to playing the "second-guessing Irex game"

But doesn't a SDK imply that there's somewhere to run the software created with it...? While I could write a Universal Reader E-reader for the Iliad its a bit academic until I can run it somewhere...

Well, you have Antarctica emulation environment. I can tell you of a lot of projects for mobile phones in the universities done only with simulators because of the lack of money to buy the full system.

And the source code was released after 2.7.1...
But it was explicitly told that the hole was going to be patched. This is different than 2.4, where no explicit announcement was done.


Not to mention that a full SDK would have the examples + docs (in an ideal world :) ) you mention.

Indeed that is important; specially to agree in the way of interaction with the user (the toolbars, the updates). But look Zaurus, or even Maemo :(

Now, to us it looks like iRex is working on closing the "holes" to prevent people from using the community-created software.

I thought Irex position is that they don't like people running untested software as root on the Illiad. Their argument is that it can result in bricked Illiads which they then have to spend time unbricking. Seems perfectly reasonable to me.

There's nothing stopping people with pre-2.7 Illiads running community based software.

They also appear to be working on a SDK that runs in user-land. And they're hinting it will be out soon.

So a bit of patience and understanding...? :D

(I do find it strange that people who would complain about certain companies leaving security holes in their software, are complaining about another one that wants to patch them :D )

@jaed

I think there is a slight misunderstanding here...the sentence you quoted out of context makes it look like I did not understand that they are trying to stop people from bricking their Iliad, but I wrote in the next sentence that we all know their real intention...

Now, to us it looks like iRex is working on closing the "holes" to prevent people from using the community-created software. Of course we understand that it is to keep us from bricking our Iliads, but what we don't understand is why we don't rather get a "legal" way to install and run those apps.

This is the full quote and I think it makes my opinion clear.
And about the patience and understanding...well, I think we have all shown a lot of that, but we get release dates that aren't met and promises that aren't held (and let's not forget the violation of the GPL), and those who would like to help (like all the developers here) get little or no chance to do so.

Matthijs seems to be a nice guy and he does try to help, but one guy occasionally giving hints just isn't what is needed here I'm afraid. I'm not for leaving security holes and making people risk their Iliad, I'm for finally prioritizing the development in a sensible way. I can't say what the b2b stuff is doing, but from a consumer perspective it doesn't make sense to add scribble to PDF when the basic functions are still far from complete and the device is slow as hell.

Most this indeed boils down to you guys being unable to view the entire picture and us being unable to show you the entire picture.
I can understand that some of the development decisions/priorities might look strange from your point of view, but they make perfect sense from ours.

@Matthijs

I don't know whether you can say anything, but if you can, can you give us a short snippet on what this b2b focus of development is all about?
What is the b2b market in your case? Is it a "reseller" market where some newspapers and ebook distributors want custom software for the Iliad that only runs their proprietary formats? Or are you selling Iliad devices to companies that use them as worktools for their employees...and why are teir needs so different from ours? Don't they need a fast and stable, paperlike device?

Thanks. I'm really sorry to see all this bickering going on here, and I'm equally sorry to be a part of it, but I guess none of us can stop it if we don't know what really is going on. We paid a lot of money for a device and of course we want to use it our way...

I cannot make any such statements here (or at our own forums), I'm sure you'll be able understand that.

Most this indeed boils down to you guys being unable to view the entire picture and us being unable to show you the entire picture.
I can understand that some of the development decisions/priorities might look strange from your point of view, but they make perfect sense from ours.

Weird... In my company the development priorities look strange from *everyones* point of view... :D

Not really since I don't see where the big secret is, but never mind. I had expected as much.
Thanks anyway.

This is the full quote and I think it makes my opinion clear.
And about the patience and understanding...well, I think we have all shown a lot of that, but we get release dates that aren't met and promises that aren't held (and let's not forget the violation of the GPL), and those who would like to help (like all the developers here) get little or no chance to do so.

Fair enough... But what do you want...? Release dates that get broken, or no release dates and uncertainty...? Ideally it would be release dates that don't get moved back, but I also want the moon on a nicely carved stick...

The Truth...^^

I know that I'm asking for the impossible, but those who don't never get anything.

The main point however is, that we are some kind of "customers second class", something we knew about from the start since the focus was b2b and that was clearly stated, but the lack of concrete information hwat is actually going on is more than a bit frustrating...but never mind.
I just hope that the community manages to compleetely hack the Iliad and make a customized OS or that iRex gives the developers enough sources and information to optimize the existing system.

After years spent dealing (or not) with Psion, I'm really amazed at the amount of support IRex is prepared to offer us. The bottom line is that they have sold us a product and we have purchased it. And assuming it isn't faulty, that's probably most of what they have to do.

Yes, the law states that they must release the GPL sources, but if they really didn't want to it would take several years to enforce.

I see writing for the Iliad as a fascinating challenge. If they provide us with access then great. If they don't then we'll find a way. If we get locked out completely by them then we'll buy another toy to play with. I compile up apps for fun. Maybe somebody will use them, maybe not. But I don't expect anything back from IRex either way.

If the dozen or so vocal users on here make any difference whatsoever to their B-B plans, then those plans are seriously flawed. The Iliad will succeed or fail regardless of what we do or say.

So... I think of anything coming from IRex as extra chocolate. :happy2:

Maybe you are right design256...how about reopening the meditation thread...^^

Hi All,

have you read the reply from Matthijs on design256's question in the iRex-Forum (http://forum.irexnet.com/viewtopic.php?t=635) ?

...
Basically to have all functionality (including hardkeys) contentLister must launch you application.
Controlling the pageBar in the situation you describe should be possible though.

The best way to make this work is by using the shell-script addon to the contentLister that is floating around...


For me this is really interesting ;)

What's the status on this pluggable ipdf? Can I download it anywhere?