We need a new web browser!!

Is anyone with me here? IE 4 just doesn't cut it anymore. Netfront is tough to get working and has a lot of display bugs. ftxBrowser is IE4 with tabbed browsing and alt-tap menus to save pictures and targets (which is about the best you can get right now).

When will the folks at Bitstream compile their MIPS release of Thunderhawk 2 for HPC 2000? Or make their ARM release work with the Jornada 72x?

What about the folks at Mozilla, who make a version of their web browser for practically every operating system in existance (including OS/2 !! ) except Windows CE?

If Microsoft has released IE 5.5 on Pocket PCs and 6.0 in Windows CE .Net 4.1 and 5.0, then why can't we get at least a version of 5.5 for the HPC 2000?

Honestly, I think it's a tad sad that one could buy a used Sony Picturebook or Toshiba Libretto in the 266-400mhz range and run Windows 98SE with better web browsing support than the Jornada 72x or the Mobilepro 790/900, for about the same price as buying a J728 or MP900.

Anyone else wanna complain with me?

I've got a Libretto 100ct, overclocked to 266MHz. it does browse nicely, even running Windows 2000... Horses for courses though.

We do need a new web browser: it's the most talked about topic on this forum already.

Problem is that a web browser is not a small monolithic piece of software that you can just have a hack at and make work: it's a large, complex piece of software engineering.

Perhaps persuading thunderhawk's authors to look at the HPC marketplace would be good, but given that we're a dying breed and that the newer browsers in CE .NET are acceptable, I think we're on a loser.

The community, if it wants a browser, will have to write its own. Eek.

I find this so intresting but yet i dont understand much about proramming itself....One thing i cant get out of my head is how a small B&W symbian os 5 or epoc 5 device can oout do a wince divice on the browser still today, Oprea 5 w/java? i've had a Diamond Mako, and on that little thing it ran great...why is this a problem to do on a WInCE HPC device? when a i small & dinky 17mhz arm7 in the Oregon sientific osaris epoc 4, $50 dollars if you can find it online...can run it as well.....what is wrong, why cant we do it, why is it such a problem?.....Steve

< start of rant >

The problem always has been with user apathy...

In the Psion community there are such a rich breed of developers over there. They weren't lulled away from the devices by the Microsoft marketing machine as was the case here.

They also have superior (I'm led to believe) development tools. Just look at how difficult it is to code for the HPC2000 API. No emulator... and even the 2.11 emulator recompiles the code into x86 and doesn't always give a true representation.

... but back to apathy.
If I had ashiney British penny for ever time I heard
"It's Microsoft's fault", "I don't want to pay for it", "I want it free", "Can I have your license", "Will you contact xxx for me", "Why should I...", "Do you have a crack for..."

Then there's the occasions when we've tried to drum up support for something... the dissaster that was the MacOS sync client springs to mind. We had one of the PPC developers say yes to writing a HPC version of a MacOS sync client.
So many people around the place asked for this. However when it came to actually doing something about it... do you know how many people contacted us?

2

One even gets the odd email now from people saying "well where can I download it?"

I could give you countless other anecdotes of when all that was asked of the community was to email in letters of support, or spend 5 £ $ € ¥ for something that a lot of people wanted. Where no one could be bothered to do anything.
This mindset simply doesn't exist in the Psion world.

I was talking to a developer about 6 nights ago - who shall remain nameless. He told me that he was bitterly disapointed with the way users had treated him and his application, they offered no moral support what so ever. He had slaved away for months learning new skills to write the program - which was something people had asked for and restored missing functionality to devices.
He had reduced the price of the application repeatedly.

I wont say what the sales figures were for it. However it certainly was not good.


irobertson, yes it is sad. That's why I think we need to get the H/PC and P/PC reintegrated again into the one platform. With a universal, mainstream, main marketing course Microsoft would be happy. The platform would then be back in the realm where all the developers are.
That would get you all the software you you ever wanted. Including new web browsers.
Although on the IE5.5 point, you are forgetting the lines that are drawn between MS PDA released. The OS seperation is not the same as with mainstream Windows.
Windows CE 3.0 HPC2000 and PPC2000 came with PIE 4.01 SP2
Windows CE 3.0 PPC2002 came with PIE5
Windows CE 4.0 and 4.1 came with PIE5 -or- IE5.5 CE Edition
Windows CE 4.2 comes with PIE5 -or- IE6 CE Edition

MSIECE is not a solution for devices that are below SVGA resolution. There is little to no point running it on a HVGA IMHO. HPC2000 will never see a full IE release because they are coded for the 4.x platform. This also goes that PPC2002 will never see one either.
The OS has end-of-life'd. As is the way with ALL PDA OS's. However as I'm sure you've seen, we just need to provide developers with the incentive to come back to us. That's all.

Phatware have, Illium are considering it. SoftMaker very much have after a wobble over PlanMaker. Just think pro-active!!

< /end of rant >

We in the h/pc world need to collaborate with those with similar interests. Agreed? Collaboration will get us much further than competition.

I seem to be a good scrounger / facilitator. With those thoughts in mind I found a site that made a standalone version of PIE for MIPS based on the ce.net PIE build. It looks intriguing to say the least. I think an updated version of PIE (IIRC thier package contains PIE 5.5) is a smashing start.

We'd need to work with them as the distributable package is something I can't make heads nor tails of. That and they deserve the credit for developing the package.

If interest is warranted I'll pass on the contact info to mods / siteowners here.

Still ,we should continue to politely badger Opera/Mozilla to compile for us. (Mozilla is currently porting for Phillips mobile devices.) A cutoff of hpc/pro (2.11) 65k colors should please a huge array of users. Has anyone thought to contact Access Japan? AFAIK, they abandoned all but ARM products. Perhaps they may allow or license a subset of Netfront for other platforms. Netfront is such a good mature product it would be an incredible starting point.

Our h/pcs have a great office suite, good financial programs, fabulous mp3 players, great video players, more games than I can master, connectivity of all sorts, and an extremely hapless browser.

Give my 690 and 6651 a browser designed for them and they'll see 5 more years service with me.

Chris, I share the frustration...

Dev tools for HPC are not that bad, but what is killing us is the lack of a decent emulator. As you probably know, my application works wonderfully on the 2.11 emulator, but refuses point blank to work in the same way on my HPC. The reason? Every OEM is left to implement routines to microsoft specifications, and everyone does things in different ways... Also, the emulator isn't "emulating", it's implementing some extra APIs and granting access to base Windows APIs. As a result, things work that shouldn't work in WinCE.

Developing for Symbian is, on the other hand, different from developing for Win32 or WinCE so at least we've not got to throw away everything we've already learned and covert over to another programming methodology!

Collaboration is the name of the game. Are there any other developers out there with inclination to do something?

Cheers,
Nick.

I saw some tools the other day to resovle those development problems, problem is since i havent a clue about devlopment, i just took a look and moved on in my software search, if i run across the site again, i'll let you all know...Steve

"He had reduced the price of the application repeatedly.
I wont say what the sales figures were for it. However it certainly was not good. "

/rant
Some of the developers though are outright idiots with their pricing.

There's a financial app I'd like to buy. $40 was the intro price 5 years ago and so it remains. Hey, MsMoney is free. I'll use it and make do. There's a PIM program I'd dearly love to have. Same deal. $80 at intro and still $80. I want to shake some sense into those developers. This is a limited market to sell into, each year more of these devices die. Each year people will be more reluctant to spend more on software than what their used hpc cost.

Good thing much of the crap is time-limited functioning demos. Otherwise I'd have wasted $50 from a certain developer for his crappy buggy not-updated in several years encryption program. When emailed about support or updates I was ignored. Why should anyone like that get even a nickel from our community?

Basic economics. If your goods don't sell your price is too high or you're not reaching your audience. Most of us are frantically scouring the net for hpc apps. If we're not buying what we find it's likely the price is out of line, there is poor customer service or it's non-existent, and a biggie, support/updates seem few and far between.
/rant

I'm not asking for anything for free. I'll pay a fair price for a product based on the price of competitive (including free) products and the value the product provides to me. However, I won't spend a nickel just because someone wrote it and expects me to pad their pockets while they neglect to support and update their goods.

Edited by wallythacker 2004-10-21 5:37 PM

Wally,

I understand what you're saying and yes there is a valid point there. However you have to ask yourself why the developer just stopped dead in his tracks in the first place.
I doubt that it was through pure rudeness.

If no one is using the application, why continue to write for it, bug fix and update it. If you're not getting bug reports back from users, how do you know what bugs there are to fix?

I have run beta's here in H/PC land before. I have been mortified at the lack response. I shant be doing any more in a hurry... and my beta's were all free!

You are correct there are some very highly priced applications out there for the H/PC. From commercial developers. Who, unlike closet developers have salaries to pay the programmers. With them it isn't done for love.

I agree that keeping programms inflated for a prolonged period of time is commercial suicide. I am also appaled at the number of developers who pull application with shareware trials floating about the internet. bsquare really got peoples backs up when they pulled out.

Their utilities have to be the most requested "where can I get" application I've ever come accross. In my humble opinion any developer looking at pulling out should either do as Utopia did and release a public key, maintain a shop on handango - lets fact it it isn't going to cost them anything to have that there and keep their bank account details up to date. Or lastly hand the application over to another active developer / site to sell it.

We here at factor wouldn't hesitate to do anything we could to keep an application in the public domain.

What about tech support you may ask... well there is an issue there. Granted. However selling the application under the premise that it is a Final release and that tech support is only available through the community / faq is a work around.
Yes it would be better if they didn't drop out. However sometimes there is nothing else that one can do other than accept it.
How many people are out there that would still wish to purchase bsquare application, knowing that it's a final release. I'd say there are a lot.

I just wanted to add a quick comment to the rest......I love it when a program cost $15-30 and 50% of the time its not stable on a continuous basis.....well you get my point....Steve

Ok, I personally have to rant myself - here it is.

Has anyone here thought of starting the development themselves? I mean, mozilla and firefox are open source! Why go after the developers to make something they don't need when you can just join the project itself or port it on your own? I mean, I personally am going to take up programming on my j720 the second I learn, and I'll try to fix some of these problems on my own. But seriousely, if something for an obsolete system has to be made, it will have to be made by the people who use it themselves. Come up with a new emulator that works, do it all. Palm OS has that kind of community, but CE doesn't seem to. If we all want something, we are going to have to do all the dirty work.

Lots of good opinions here both ways. I developed a long time ago for large business.
Totally different than developing for retail. Many of my large projects were finish/cash out/walk away.

That doesn't work for retail apps. The encryption program I mentioned hasn't seen updates in years.
I'm sure the developers of it have known for quite a while what bugs remain. Another quirk is the in/out developers.
Access Netfront 3.0 is pretty decent on my h/pc. I'm sure it would make a lot of mp/68x/72x users happy for
quite a while. 3.0 is already coded for ppc ARM/SH3/MIPS. How tough would it be to tweak and compile for 2.11?
But what did Access do? Dropped everything but ARM in the 3.1 release. Ugh.

In theory one can still buy 3.0 (accordiing to access) and hack it for their MIPS but I couldn't find it for purchase
anywhere online. (I didn't look all that hard, if anyone finds it cheap, $20US or so then go for it. A small hack
and you're good to go. It's way better/slightly worse then PIE. ) Access would do the world a great service
offering 3.0 to the public for free or super cheap.

I hope small developers are motivated by craftsmanship and not money. They won't get rich if money's their only
motivation to code.

Anyway,

It is easily said and I'm sure it has been flashed through more than one brain around here... but...

Hands up all the high level C++ programmers

*tumble weed rolls by*


Anyone who wants to start can get all the bits from here:
http://www.hpcfactor.com/developer/

Admitedly a more powerful approach to eVT is to have Visual Studio 6 and the CE Tool Kits... but neither are free.

Yes, I do understand that it would be hard and take a lot of work to develop software for this system, but think about other examples out there. I hope I don't annoy others if I relate to linux a lot, but it is a good example. You have the hardware, get a bunch of people together, and develop for it! All that really needs to be done by the HPC community is to create a programming system that can be used on the HPC and release it under the gpl license. Once that is in, people will be able to develop with nothing in the way or even having to pay anything, and maybe this almost-gone system will get a new life from open source developers. I know it can be done, but if everyone shys away, this will never happn, and I will, like I did before with other operating systems, watch it fade away to never even having use again. I actually thought that that ha happened for these systems until I found this forum, though small, gives me and others the chance to figure out stuff with others. Come on everyone, this can be done, I know it!

-I am sorry that I wrote an essay hehe but I really do want to get the point across of what can really be done if we work together on this.

We can talk about this til we're blue in the face.

Mozilla/Firefox is ah, to put this politely, rather large. The source is a 31.6MB download. Then, to build, you need cygwin and Visual Studio, or you can try with GCC. Oh, and you also need netscape's wintools.

I have not even tried building mozilla/firefox on Win32, let alone porting to WinCE. I am interested, but ultimately need some reassurance that firefox will be performant on a 16MB WinCE machine before even attempting anything to do with it.

Porting mozilla is a huge effort, and is realistically speaking well beyond me. We would need a team of 5 developers minimum (of which I can be 1) who could carve up the source and start to understand it to even understand what would have to be done to make FF run on WinCE.

Dillo, as I have said before, looks our best bet. This is intended for Linux running GTK+. There is a Win32 port of GTK, but guess what? No WinCE port... Porting Dillo would IMHO be best achieved by porting GTK, then attacking Dillo. We'd need to compile with GCC, GLIB and all that sort of stuff too. A couple of projects exist to port GCC to the PocketPC, but nothing works on the HPC yet. So that's more work.

Don't get me wrong: I'm all for this.

However, time for reality.

No-one else is going to port a web browser to Handheld PCs. Not even people who already have a commercial PocketPC web browser.

Why?

We, the user base, are too small, and we don't like paying for things. "Firefox is free, so why should I pay for this" would be the mantra as the application was ripped off mercilessly left, right and centre. Even if everyone did pay for the application, the vendor would have to tool up to compile and support the application. First challenge, where do they find hardware that is supported by the manufacturer that is running HPC Pro, HPC 2000, ... ?

That's enough spleen vent from me.

If people want a new browser, we need C or C++ developers, preferably either with experience of Linux programming (Dillo) or WinCE experience.

There's one here.

Anyone else?

Cheers,
Nick.