Slashdot Mirror
When Cellphones Become Webservers
An anonymous reader writes "Nokia is experimenting with turning mobile phones into webservers, according to an interesting article on Linux Devices. Nokia has ported the Apache webserver and a few other software modules to the Symbian OS that runs its phones, but there shouldn't be any barrier to adapting the technique to Linux mobile phones, since it all appears to be released under Linux-friendly open source licenses. Just think of the possibilities of having a webserver in your pocket!"
I won't be satisfied until Adobe ports Photoshop to cell phones. Now we're talking.
Vibrating cell-phones that vibrate based on hits to the server!
I have mod points and I am not afraid to use them.
Just imagine the battery life of your cellphone after a slashdotting! :-p
Is that a web server in your pocket, or are you just happy to see me?
But then with opensource, I can figure anything out... like using Skype to make my calls while my faithful website viewers are still able to browse my ever-so-important website in my pocket.
At least now when the police raid your torrent server, everyone can call and tell them they're pissed.
Swedish Police seize Pirate Bay server decoys, Real Server Escapes In Man's Pocket.
What has it got in its pocketses, my love? Tricksy little serverses, sneaking awayses from us!
Imagine one of those phones being slashdotted, the sheer amount of radiation and electricity involved would cause the said nerd to spontainiously combust in an excellent display of nerdyness...somewhat like a moth to fire.
The possibility of paying massive bandwidth fees to Cingular, for example.
GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
To be able to run distributed computing applications like BOINC on your cellphone when it's not in use. It would suspend the activity when the battery charge reaches a user-defined limit.
You could crunch units at night while your phone is charging.
I just don't get it.
A web browser, I can see the use of (though currently most non-text-only pages look like crap on tiny cellphone screens, and even text-only doesn't look great). An email client, sure. A terminal emulator (aka "telnet/ssh client" for you whippersnappers) so I can connect to and manage a remote web server (if absolutely necessary - see point 3 below), yuppers.
But an actual web server?
First, my phone has an okay battery just sitting idle, but in actual use it dies within a few hours. Running a web server implies basically continuous use, so the thing would end up always on a leash to either a car or AC outlet.
Second, although I have pretty good cell coverage in my area, I do still drop the occasional call. Do we really want to add a http error code, "604: server drove into a tunnel"? (And yes, I do realize that would probably come back as a 503... Just a weak joke).
Third - I would not want to use a phone's crude keypad to try to maintain a web site. Even if I bought into the rest of the idea, I could see myself realistically connecting to my phone remotely from a real PC to do any updates or maintenance.
I just don't see the point. This smells like a solution in need of a problem, IMO.
Just think of the possibilities of having a webserver in your pocket!
... uh, no. Aaand not that. Hmmmm....
Ok, hmmm, let me think
*chirp* *chirp* *chirp*
OK, you got me - what are those possiblities?
Trust the Computer. The Computer is your friend.
I don't see much use in this... Ususally (at least in GSM GPRS and UMTS 3G networks) the phones are behind one or two NATs. That is, all packet data users of an mobile operator are seen to the internet as coming from the host gprs.mobile.operator.com, or the like. You cannot directly connect from the internet to a specific mobile phhone's IP address, regardless of the existence of a mobile web server there.
NATting is partly done to protect the mobile users from excess traffic. Imagine someone pingflooding your mobile's IP address, and you paid for data packet traffic by the kilobyte! :)
I see this webserver porting more as an technology demo from Nokia's part: "Hey look how cool our phone operating system and programming platform is!", instead of being a real, useful application.
It reminds me of 3 years ago when I first got a Linux-based Sharp Zaurus. I could purchase a GSM/GPRS card to get it acting as a cell phone. And I loaded it with wifi and packages so it acted as a Samba server, an Apache server, a MySQL server, a VNC server, etc. Nice geek attraction but for practicality's sake the usability was pretty poor. Small devices aren't geared to be resource hogging servers. They are optimized to be thin clients.
You know how much it sucks to have to configure your phone's settings--not just time and date, but network preferences, calendaring options, notifications, and all the rest--on the phone itself? Now imagine firing up Safari and simply browsing to your phone's configuration page, where everything's explained in full sentences in a format human beings can read, not crammed into 1 square inch at 288 dpi, and where you don't have to press twenty nubby little buttons every time you want to change one setting.
This could be one potential use for a webserver on your phone. Given the complexity of your typical cellphone, I'd be glad to configure it through an interface that sucks a little less.
And now, a PSA from David Lynch.
With all the hype of rapid development frameworks (Ruby on Rails, TurboGears, etc) it's easier than ever to make web applications, for yourself or someone else. It's also damn easy to install them. Only problem? They require a web server.
Having a webserver on your cellphone, even if it's only accessible to you, is extremely useful. You can build your own truly cross-platform applications without having to worry about crazy microjava doodie.
In terms of power consumption, why would it have to be continuously active? It can have a "sleep" mode just like anything else on a cellphone does. It's not like your phone has a continuous open line to someone. When you finish talking to someone, it goes into a sleep mode and waits for the next call. A webserver could work the same way -- when you use it, it fires up. When you stop using it, it takes a nap. Both, you and your battery, are happy.
I, for one, welcome our Cellphone-hosted website overlords.
- shazow
The first thing I would do is have the phone update to a page, wherethehellismyphone.com. I would have this constantly updated with rough GPS coordinates. I'm not sure what phones can tell you their rough GPS coords, though. But it could give you all sorts of useful information in case the phone is lost or stolen -- what it last saw, where it was last used, etc.
I mention this because I recently picked up this laptop, and one of my first plans is to get a GPS card installed in it. I'll have it running something netstumbler-like, and if it's lost or stolen, it will do its best to log in and upload the GPS coords to wherethehellismylaptop.com. So, if my laptop is lost or stolen, and the thief leaves it turned on while passing through any open wifi or going online in any way, presto. I could have the site have a Google Maps thingy that shows me where it was most recently spotted and when.
This doesn't even require the GPS card -- any information you can have the device update you with is useful. It could tell me what the person was last looking at, what pages they're frequenting, etc. Get their name from their MySpace page and have the police show up at their door. Letting mobile devices act as servers opens up a lot of these possibilities, including making them easy to use as James-Bond-type spy/bug gadgets and taking a big step in the direction of useful remote presence.
Of course, wherethehellismylaptop.com would require a very secure login if you want any privacy, ever.
xkcd.com - a webcomic of mathematics, love, and language.
Actually, we do have always-on cellphones when it comes to TCP/IP. Both of the major international standards, GSM and IS95 (well, ok, the latter isn't that major, but it's #2 so it gets a mention) have always-on TCP/IP packet data. GSM has GPRS and EDGE, and the 3G variant, UMTS, also has packet switching as a basic service.
For the people rubbishing this, I have one thing to say: WTF is wrong with you people? Why do you short-sighted twits appear the moment anyone mentions a technology combination you've not thought of?
This is just the implementation of a protocol. No, hosting your blog, let alone a major ecommerse site, on a cellphone is probably silly, but if you're looking at implementing some base services, especially for something like telemetry, HTTP is an obvious choice if you have the hardware on the remote end that supports it.
HTTP is well supported in Java, .NET, Python, Perl, and a host of other languages, so the software that runs "back at the base" becomes far simpler to implement if you're going to be accessing information via HTTP, rather than convoluted customized protocols based upon UDP or SMS. What do you think's easier? A call to the HTTP library to fetch http://mobilstation7.intranet/cgi-bin/getcurrentte mperature.exe or custom formatting some UDP packet with a custom designed library and sending that?
Is the objection that HTTP has too much overhead? A bare-bones, stripped down, Apache isn't that large, and look at what you're talking about running it on. A modern mobile phone typically has several megabytes of RAM and 8-16Mb of flash, plus bluetooth or USB interfaces. If it didn't, the camera on it wouldn't work.
A mobile phone isn't a dumb handset, it's a moderately powerful computer that acts as a mobile terminal in a cellular network. You may use yours purely for voice applications. That doesn't mean the only application for this remarkable technology is voice driven. Telecommunications is a versatile instrument, and anything that makes certain types of application easier to implement is to be welcomed, not laughed at.
You are not alone. This is not normal. None of this is normal.
It could be great for engineers like me that deploy a lot of short-term and long-term measurement systems (noise/vibration/temperature/wind speed/etc) and want to make the data available in real-time to interested parties (e.g. a local community).
Currently, the only way to disseminate this info is:
I would love to just be able to hook up a cellphone to the data logging unit, and just point people to www.city-noise-monitoring.org/site1. Yeah I know, niche application.
The only issues I see:
I for one will be watching out for this.
I feel it's nuts.
If you think grammar is unimportant, just think what the absence of an apostrophe would do to the above sentence.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
However,
Which makes me ask, "In that case, why the hell would I want it?"