1. WAP really isn't documented anywhere of what acutally works and what doesn't. Not to mention that every phone maker out there is using something different, It isn't really WAP, it is the bastard first child of what will become WAP in the next 6 months. There is limited support for full form functionality. And the emulators don't work consistently, thus, the phones don't work consistently, get my drift. For the Mr Rodger's neighborhood people," Can you say Cluster F*CK!"
True, though it is a pain to design a website that uses the latest HTML tricks that looks good on all browsers too.
2. Speed burns baby! WAP, at best can handle 9600. Whoo! now that is a screaming technology. But that is what it was built for, recieving text, not graphics, not games, text. I-mode however has been built for speed. In Japan they are playing network games over these things.
The article confused WAP with the networks it can run over -- WAP can run over packet switched networks just as well as IMode can. It just isn't deployed in fast networks today because the only fast wireless networks today are in Japan, where the near monopoly NTT DoCoMo has chosen IMode.
When WAP decided to come out, it thought, hell, we don't need no stinking http protocol, we will invent our own called WAP. yeah, great idea guys, did you forget to mention that you are re-inventing the wheel here or were we supposed to see something great and new.
WAP is forced to be diffrent by the crappy networks it must run over. It is not re-inventing the wheel, it is making a wheel that actually works.
Ever hear of Imode? It's likely to blow the doors off the WAP standard. Practically exponential growth here in Japan.
Imode is pretty awesome, but it is specific to NTT DoCoMo (the dominant cellular carrier in Japan). While DoCoMo is a powerhouse, I don't see its standard being able to take over the world. In fact, DoCoMo is a member of the WAP forum!
The rest of the world is behind Japan in terms of wireless technology, so the rest of the world will start with WAP (which they can and have deployed today) and WAP will grow with their networks.
My basic complaint is the premise. On the one hand we see a whole new type of device with legions of people trying to figure out how to make efficient GUIs while conserving either display space, or storage, or whatnot with WAP...and on the other hand we have multi-zillion dollar companies building infrastructure and vastly powerful processors, that will render the need for "efficiency" as irrelevant as my 2gb hard drive.
My prediction is that we are going to Moore's Law WAP to death in short order ("I'd like 'The Patently Obvious' for $400, Alex")
WAP will become obsolete. It'll actually converge with internet technologies as a whole. But that doesn't mean it is useless now. We're a long way off (years) before the wireless networks can deliver the same amount of bandwidth that, say, a DSL line delivers to people's home today. It doesn't make sense for content companies to just sit and wait for WAP to fade away -- the potential size of the WAP market is too huge.
I'm a web application developer in Canada, and what I'm noticing is that investor types jump out of their chair at the mention of WAP, but it's hard to find a soul that knows what's actually going on with it. I would suspect that it's largely because there is not a single phone/service combination in Canada at least that allows you to type in a url and go to it - wml or not. You're restricted to whatever boring content Bell gives you.. do we really need more sports scores and horoscopes?
There are several things causing that situation:
- Right now there just isn't very much random WAP content available, so there is no point in a "goto URL" feature.
- The cellular carriers don't like the idea of users going just anywhere -- that diminishes the leverage they have with content providers.
- From a usability standpoint, most cellular phone users aren't interested in typing in a 30 character URL with their keypad.
But eventually there will be enough WAP stuff out there that omiting the "goto URL" feature will lose the carrier some customers. It'll come.
Sorry, but to incite developers to use WAP it takes more than a free dev. kit, I think.
Actually, what will incite developers to use WAP is the existence of phones and gateways that support it!
The people who decide whether this happens are not geeks reading Slashdot, it is the big wireless carriers who have invested huge amounts of money in their networks. Why do they choose WAP? Quite simply there is no other set of protocols that more adequately addresses the issues involved in deploying applications over the wireless networks they have spent all their money on.
WAP is simply not an open standard. We already have a technology for such applications, whjich is proven, open and runs pretty well : HTTP.
HTTP is not analagous to WAP. WAP is more like TCP/IP + HTTP + HTTPS + HTML + a browser operating environment + Javascript, but all designed in such a way that it can run over all the disparate kinds of networks that exist in the world today. In fact, WAP does use HTTP, but that is just a small part of the picture.
As for WAP not being an open standard -- there are real efforts being made to make WAP and internet standards converge (e.g. both WML -> XHTML and HTML -> XHTML).
This will happen at about the same pace as it takes the wireless carriers to convert their networks into beasts that look a lot more like the internet at large (similar bandwidth, IP based, etc.).
...and there is no way to make it faster because it's in java? I mean, it may be fine for windows users who believe that Unix boxes must have user interface as slow as Microsoft telnet makes them look, but for any real use X server must be fast and be capable of using graphics card's acceleration for everything that card allows to accelerate.
Even if it is slow, machines keep getting faster. The faster machines get, programs written in these higher level languages get more and more viable.
So I see this as more of a theoretical interest than a real practical X server.
I think at some point we'll have a watershed moment where the programming efficiency gained by writing stuff in Java (or similar) outweighs the performance problems (which diminish over time). The Java programs will just be more featureful, more robust, and will be able to adapt to feature requests more quickly. We're already seeing this with more and more significant apps written in languages like Python.
Their crawler (FAST-WebCrawler/0.3) was not very nice when it blasted through my site. The general guideline is that a crawler should grab one page a minute. Their crawler grabbed multiple pages a second even when going through a bunch of pages generated by CGI.
NSI is whistling through their ass here. MAPS RBL is a purely voluntarily blacklist. If someone does not wish to have their mail to/from NSI blocked as a result of the RBL, it will take approximately five seconds for them to disable RBL filtering on their mail server/routers, or provide an exception for NSI's IP address space.
MAPS RBL is painful for those who have ISPs that use it. They have no way to opt out of the "service" their ISP is providing. Sure, they are "voluntarily" using that ISP, but when you have to switch ISPs to avoid spam protection software you are inconvenienced to a much greater degree than a few spams a day.
They're refering only to the software. As in, Apache is a web server.
They say it needs only 750 bytes of ROM and 28 of RAM which astounds me, but whatever...
What Emware has done is totally believable. It sounds like the server is communicating with the browser with those "microtags" which sound like ~2 byte integers to me. Their server doesn't do TCP/IP, just a "simple serial protocol" which also implies very little code. And the 750 byte ROM estimate probably doesn't include the code to make the device actually do anthing when the browser tells it do -- it is just the framework for sending and receiving commands.
In other words, it sounds like what is in the device is just a simple framework for communication over a serial line (simpler, even, than a gdb sub!). Emware's business model has got to rely on their off-device technology or they are doomed.
They also claim to have trademarked the term
Dynamic Expansion
which I find really funny. They even want to patent it.
Making money is the next big problem facing the whole open source thing as a concept. Once an OSS probject gets "swallowed" by commercial interests, the willingness of folks to just donate their time on it dwindles. It ceaces to become a hobby and instead becomes a job. Look at Mozilla -- who wants to donate their free time to that when there is already an army of people paid to work on the code? Look at sendmail -- now that it is sendmail.com, who outside of that company is ever going to really work on the program? Same thing for tcl/tk and scriptics, etc. Problem is, once this happens, a lot of the advantages of the open source development model go away.
So, I just hope IBM and Apache figure it all out for the rest of the planet.;-)
Color displays require more power. More RAM requires more power. More power means shorter battery life. The #1 reason I love my Pilot is its battery life (I get several months out of mine). So I won't be upgrading from my "Palm Pilot Personal" any time soon.
Here in Sweden when we recruit new software engineers direct from college they show no particular interest in Win32. Instead they are fluent in the UNIX/Linux environment. I have also got the impression that this distaste for Win32 is valid in most countries.
When I was in school 4-5 years ago the draw to Linux was the ability to have a similar development environment at home that was at school (Unix). In the United States, at least in the University of California school system, I've noticed a good number of Windows machines infiltrating the CS departments. Can you believe that they'll set up a WinNT box running a Hummingbird Exceed X server as a terminal instead of a Unix box?
I wonder if they are doing this as a response to complaints from industry that new grads don't know anything about Win32.
On a side note, my new employer is having a hell of a time finding developers who want to work on Linux.
Microsoft just buys companies like Hotmail and LinkExchange after they are already established. They aren't stupid and so they don't force them to re-deploy all their software under a different OS (WinNT, whatever) for no good technical reason.
Look at all the portable Linux projects right now, like Linux for the Palm platform, etc. It can be run in a small memory footprint, if memory srves me correct (2.0.x could run in 4MB or so). I believe WinCE needs at least 8MB (or possibly more) to run.
And if you're looking at set top boxes, you're looking at things like playing multimedia files, heavy web browsing, heavy graphics, java, etc.
With all that, you need a lot of RAM to just get a usable system. It is clear that for anything that doesn't run off a battery, the difference between 2 and 4 megs ram (or even 8 megs) is not all that significant.
Quite glaringly. It mentioned lack of plug n play support. Where'd this come from?
I'm running 2 PNP cards in PNP mode in one machine at home, a sound card and a lan card. In another box I have a pnp sound card. (All are ISA PNP cards). They work fine with an out of the box RedHat 5.2 install.
That's not all there is to PnP. There is no PnP support for monitors as far as I know. If you stick a new card in your machine, does Linux detect it and automatically install the correct driver? That's the kind of thing people coming from Win32 will expect when you say "PnP."
San Francisco is, relatively speaking, not a high tech place. No major peering points, no major co-lo facilities. Limited DSL. No cable modems. I think they gerrymandered the Peninsula onto the San Francisco ledger just so they could put the cute cable car photo on the page.
You're talking about network infrastructure which wasn't factored into their analysis at all. They were trying to go for the city with the biggest sociological and economic investment in "the net" in general, not the city with the biggest pipe.
Slashdot Mirror
1. WAP really isn't documented anywhere of what acutally works and what doesn't. Not to mention that every phone maker out there is using something different, It isn't really WAP, it is the bastard first child of what will become WAP in the next 6 months. There is limited support for full form functionality. And the emulators don't work consistently, thus, the phones don't work consistently, get my drift. For the Mr Rodger's neighborhood people," Can you say Cluster F*CK!"
True, though it is a pain to design a website that uses the latest HTML tricks that looks good on all browsers too.
2. Speed burns baby! WAP, at best can handle 9600. Whoo! now that is a screaming technology. But that is what it was built for, recieving text, not graphics, not games, text. I-mode however has been built for speed. In Japan they are playing network games over these things.
The article confused WAP with the networks it can run over -- WAP can run over packet switched networks just as well as IMode can. It just isn't deployed in fast networks today because the only fast wireless networks today are in Japan, where the near monopoly NTT DoCoMo has chosen IMode.
When WAP decided to come out, it thought, hell, we don't need no stinking http protocol, we will invent our own called WAP. yeah, great idea guys, did you forget to mention that you are re-inventing the wheel here or were we supposed to see something great and new.
WAP is forced to be diffrent by the crappy networks it must run over. It is not re-inventing the wheel, it is making a wheel that actually works.
Ever hear of Imode? It's likely to blow the doors off the WAP standard. Practically exponential growth here in Japan.
Imode is pretty awesome, but it is specific to NTT DoCoMo (the dominant cellular carrier in Japan). While DoCoMo is a powerhouse, I don't see its standard being able to take over the world. In fact, DoCoMo is a member of the WAP forum!
The rest of the world is behind Japan in terms of wireless technology, so the rest of the world will start with WAP (which they can and have deployed today) and WAP will grow with their networks.
My basic complaint is the premise. On the one hand we see a whole new type of device with legions of people trying to figure out how to make efficient GUIs while conserving either display space, or storage, or whatnot with WAP ...and on the other hand we have multi-zillion dollar companies building infrastructure and vastly powerful processors, that will render the need for "efficiency" as irrelevant as my 2gb hard drive.
My prediction is that we are going to Moore's Law WAP to death in short order ("I'd like 'The Patently Obvious' for $400, Alex")
WAP will become obsolete. It'll actually converge with internet technologies as a whole. But that doesn't mean it is useless now. We're a long way off (years) before the wireless networks can deliver the same amount of bandwidth that, say, a DSL line delivers to people's home today. It doesn't make sense for content companies to just sit and wait for WAP to fade away -- the potential size of the WAP market is too huge.
I'm a web application developer in Canada, and what I'm noticing is that investor types jump out of their chair at the mention of WAP, but it's hard to find a soul that knows what's actually going on with it. I would suspect that it's largely because there is not a single phone/service combination in Canada at least that allows you to type in a url and go to it - wml or not. You're restricted to whatever boring content Bell gives you.. do we really need more sports scores and horoscopes?
There are several things causing that situation:
- Right now there just isn't very much random WAP content available, so there is no point in a "goto URL" feature.
- The cellular carriers don't like the idea of users going just anywhere -- that diminishes the leverage they have with content providers.
- From a usability standpoint, most cellular phone users aren't interested in typing in a 30 character URL with their keypad.
But eventually there will be enough WAP stuff out there that omiting the "goto URL" feature will lose the carrier some customers. It'll come.
Sorry, but to incite developers to use WAP it takes more than a free dev. kit, I think.
n tProblem/one/index.html
Actually, what will incite developers to use WAP is the existence of phones and gateways that support it!
The people who decide whether this happens are not geeks reading Slashdot, it is the big wireless carriers who have invested huge amounts of money in their networks. Why do they choose WAP? Quite simply there is no other set of protocols that more adequately addresses the issues involved in deploying applications over the wireless networks they have spent all their money on.
http://www.freeprotocols.org/wapTrap/WapShortPate
WAP is simply not an open standard. We already have a technology for such applications, whjich is proven, open and runs pretty well : HTTP.
HTTP is not analagous to WAP. WAP is more like TCP/IP + HTTP + HTTPS + HTML + a browser operating environment + Javascript, but all designed in such a way that it can run over all the disparate kinds of networks that exist in the world today. In fact, WAP does use HTTP, but that is just a small part of the picture.
As for WAP not being an open standard -- there are real efforts being made to make WAP and internet standards converge (e.g. both WML -> XHTML and HTML -> XHTML).
This will happen at about the same pace as it takes the wireless carriers to convert their networks into beasts that look a lot more like the internet at large (similar bandwidth, IP based, etc.).
Nokia is in Finland. They started out making boots and other things over 150 years ago. ;-)
...and there is no way to make it faster because it's in java? I mean, it may be fine for windows users who believe that Unix boxes must have user interface as slow as Microsoft telnet makes them look, but for any real use X server must be fast and be capable of using graphics card's acceleration for everything that card allows to accelerate.
Even if it is slow, machines keep getting faster. The faster machines get, programs written in these higher level languages get more and more viable.
So I see this as more of a theoretical interest than a real practical X server.
I think at some point we'll have a watershed moment where the programming efficiency gained by writing stuff in Java (or similar) outweighs the performance problems (which diminish over time). The Java programs will just be more featureful, more robust, and will be able to adapt to feature requests more quickly. We're already seeing this with more and more significant apps written in languages like Python.
In my mind, it legitimizes OSS more than it reflects on the founders' desire for material possessions.
Yep, everybody has to eat and it is great when being able to eat is in alignment with developing software that is useful to a lot of people.
Their crawler (FAST-WebCrawler/0.3) was not very nice when it blasted through my site. The general guideline is that a crawler should grab one page a minute. Their crawler grabbed multiple pages a second even when going through a bunch of pages generated by CGI.
MAPS RBL is painful for those who have ISPs that use it. They have no way to opt out of the "service" their ISP is providing. Sure, they are "voluntarily" using that ISP, but when you have to switch ISPs to avoid spam protection software you are inconvenienced to a much greater degree than a few spams a day.
They say it needs only 750 bytes of ROM and 28 of RAM which astounds me, but whatever...
What Emware has done is totally believable. It sounds like the server is communicating with the browser with those "microtags" which sound like ~2 byte integers to me. Their server doesn't do TCP/IP, just a "simple serial protocol" which also implies very little code. And the 750 byte ROM estimate probably doesn't include the code to make the device actually do anthing when the browser tells it do -- it is just the framework for sending and receiving commands.
In other words, it sounds like what is in the device is just a simple framework for communication over a serial line (simpler, even, than a gdb sub!). Emware's business model has got to rely on their off-device technology or they are doomed.
They also claim to have trademarked the term which I find really funny. They even want to patent it.
Making money is the next big problem facing the whole open source thing as a concept. Once an OSS probject gets "swallowed" by commercial interests, the willingness of folks to just donate their time on it dwindles. It ceaces to become a hobby and instead becomes a job. Look at Mozilla -- who wants to donate their free time to that when there is already an army of people paid to work on the code? Look at sendmail -- now that it is sendmail.com, who outside of that company is ever going to really work on the program? Same thing for tcl/tk and scriptics, etc. Problem is, once this happens, a lot of the advantages of the open source development model go away.
;-)
So, I just hope IBM and Apache figure it all out for the rest of the planet.
Color displays require more power. More RAM requires more power. More power means shorter battery life. The #1 reason I love my Pilot is its battery life (I get several months out of mine). So I won't be upgrading from my "Palm Pilot Personal" any time soon.
Here in Sweden when we recruit new software engineers direct from college they show no particular interest in Win32. Instead they are fluent in the UNIX/Linux environment. I have also got the impression that this distaste for Win32 is valid in most countries.
When I was in school 4-5 years ago the draw to Linux was the ability to have a similar development environment at home that was at school (Unix). In the United States, at least in the University of California school system, I've noticed a good number of Windows machines infiltrating the CS departments. Can you believe that they'll set up a WinNT box running a Hummingbird Exceed X server as a terminal instead of a Unix box?
I wonder if they are doing this as a response to complaints from industry that new grads don't know anything about Win32.
On a side note, my new employer is having a hell of a time finding developers who want to work on Linux.
Microsoft just buys companies like Hotmail and LinkExchange after they are already established. They aren't stupid and so they don't force them to re-deploy all their software under a different OS (WinNT, whatever) for no good technical reason.
Got the stock, kde, and gnome versions pre-compiled. Lots of dock apps too.
;-)
I think the default wm is still twm though!
They don't mention how much power the thing sucks down. If it is substantially more than current technology, it won't catch on.
Look at all the portable Linux projects right now, like Linux for the Palm platform, etc. It can be run in a small memory footprint, if memory srves me correct (2.0.x could run in 4MB or so). I believe WinCE needs at least 8MB (or possibly more) to run.
And if you're looking at set top boxes, you're looking at things like playing multimedia files, heavy web browsing, heavy graphics, java, etc.
With all that, you need a lot of RAM to just get a usable system. It is clear that for anything that doesn't run off a battery, the difference between 2 and 4 megs ram (or even 8 megs) is not all that significant.
Quite glaringly. It mentioned lack of plug n play support. Where'd this come from?
I'm running 2 PNP cards in PNP mode in one machine at home, a sound card and a lan card. In another box I have a pnp sound card. (All are ISA PNP cards). They work fine with an out of the box RedHat 5.2 install.
That's not all there is to PnP. There is no PnP support for monitors as far as I know. If you stick a new card in your machine, does Linux detect it and automatically install the correct driver? That's the kind of thing people coming from Win32 will expect when you say "PnP."
San Francisco is, relatively speaking, not a high tech place. No major peering points, no major co-lo facilities. Limited DSL. No cable modems. I think they gerrymandered the Peninsula onto the San Francisco ledger just so they could put the cute cable car photo on the page.
You're talking about network infrastructure which wasn't factored into their analysis at all. They were trying to go for the city with the biggest sociological and economic investment in "the net" in general, not the city with the biggest pipe.