Intel's 'Personal Server': The Handheld Killer?

markbaard writes "Intel is developing a wireless, pocket-sized personal server that may replace laptops and PDAs altogether. The 'personal server,' which is being developed at Intel Research by ubiquitous computing wizard Roy Want, is the size of a deck of cards, half the weight of an iPaq, and has no i/o, no screen, and no peripherals. The device never leaves its user's pocket or handbag. Pictures of the personal server and the story are at baard.com."

Comments

How can it replace a PDA?

[Anonymous Coward]

It has no screen, how would you do anything with it?

Re:How can it replace a PDA?

[AndroidCat]

I see a niche for "box booths" that provide a keyboard, screen and connectivity for people on-the-go.

Re:How can it replace a PDA?

[netsharc]

Well, to quote the site and the poster of the article, it has no I/O at all. Gee, a server no one can store information in, and no one can get information out of, get yours now! Obviously it has I/O in the form of Bluetooth and/or Wi-Fi, and what a cool idea it is, as if it's a natural extension of your brain as storage. Imagine going to a job interview and just showing up, sitting down, and have your resume show up on the interviewer's computer. I'm sure the folks at the RIAA will be cringing with horror tomorrow morning. With a sane security system, maybe you can have publically accessible folders, and personal folders, and imagine being able to get MP3s just by walking around the office/campus.

It won't kill the PDA, but if will probably transform it to just an information viewing device.

Probably won't replace handhelds...

[juhaz]

I see this more like addition, or update (if it's done right, if it's NOT done right I see it as paperweight), than replacement.

Sure, it would come handy to automatically have your portable computing device wirelessly "hijack" vastly superior input and output capabilities of bigger computer whenever you're in position to use one, but I'm not convinced web server is good enough for GUI of such device.

And it would be neat for it to still have screen and input device of its own (they would be turned off for power saving most of the time), for use where there just are no desktop systems for borrowing, in such situation, this system is 100% useless.

'bout time

[PaddyM]

But I'll believe it when I can buy it. And it works on linux out of the box.

is replace a misnomer?

[whiteranger99x]

Intel is developing wireless, pocket-sized personal server that may replace laptops and PDAs altogether.

The personal server mounts on any PC that can recognize wireless devices: "Any computer becomes your computer," said Want.

Ok, I'm a little confused, are they saying that this will replace laptops and PDAs altogether or are they saying that this is merely a more flexible means of storage that can be accessed from other PCs or PDAs?

Re:is replace a misnomer?

[zbuffered]

This is just a hard disk on a radio link

no no no, this is a server on a radio (wireless) link. It won't replace your PDA, but rather change future PDAs, which will be designed with this in mind. If you don't have to worry about storing data in your PDA, but only viewing and manipulating data, it changes the makeup of your PDA. You might even have your "PDA" just be a dumb terminal with the 'personal server' doing the actual work.

Also they talk about integrating cellphones in it, so that you can have a bluetooth earpiece and your "phone" would be in your backpack or wherever. Although I guess you'd still need your "pda" (dumb terminal or whatever) to dial the number, or make better voice recognition software so that you could dial without buttons. Or put buttons on the server or...

I'm excited about this.

Wardriving in the subway!

[exhilaration]

Let's see how long it takes our enterprising youngsters to crack the security on these things! If they become ubiquitous, then the average user will use them to transport data of great value; after all, what better place for ALL your e-mail, and ALL your documents that in your pocket? Imagine being able to break into one of those things from a laptop - or perhaps another unit reprogrammed to attach to all nearby units and grab anything available.

Sounds like fun!!!!!

Biometric authentication?

[exhilaration]

My brother just suggested that biometric authentication - probably fingerprinting - would be a useful feature on these devices. It might also be a good way to introduce biometric technology to the general public.

Security

[Ghetto_D]

What kind of security features/ecryption would be implemented on this device? There have been enough flaws oncovered with WEP that this sounds like a bad idea to me. ANY computer with 802.11 access being able to connect to my portable hard drive? And how would I know since there isn't even a display?

Next Big Thing?

[rufusdufus]

Perhaps the editor's did take a little licence when they decribe this as a handheld killer. This particular device? Probably not.

This particular paradigm? Hey maybe.

PDAs and laptops suffer from the "kitchen sink" problem. In order to please a sufficently large number of people, all sorts of features have to be included, and in turn, compromises made. Some people want color, others want audio, some keyboards some need a touch pen. In order to compete buttons are placed in random places on pdas so they can work like a dictaphone. All sorts of weird things that I don't want just to please some niche.

And adding all those features raises the cost of the device.

This idea of modularizing components wirelessly might be a really great idea. With this paradigm, I don't have to fit the CPU and hardrive into an ergonomic shape suitable for pen input. Also, your hand is relieved of the extra weight of those components.
So just like desktops, you might buy your "monitor" from a totally different manufacturer than the cpu. You like Sony's egonomics but compaq's performance? Why can't you have both?
Also, with this model, you don't need all the extra proprietary gobblygook to interface to your desktop.

Sounds like a good deal to me.

Too Small, Too Slow Net, what standards?

[billstewart]

The 1GB Microdrive is simply too small; an iPod drive is much closer to adequate. The reason I'd want to use one of these things instead of hauling around my clunky laptop is so I can carry my entire Microsoft Outlook work email system (needs about 2GB) plus my personal email (needs about 100MB :-) plus the files I've been working on in the last year (another ~1GB, much smaller if I'm constantly sorting down to the files I've used in the last week or month, which is way too time-consuming for a labor-saving device.) That way I could get by with a thin client at work (~$200 plus monitor) and my own PC at home.

A 400MHz XScale is just fine for this kind of thing. But while Bluetooth is good for some applications, it isn't fast enough for many others; it's like an 0.5X CD-ROM speed, though it's certainly good enough to drive the headphone audio. 802.11 could be fine (though it tends to be a power hog, suggesting the need for an on-off switch...), though it's tempting to recommend a simple ethernet jack on the side instead, which has the advantage of working in places where people are correctly paranoid about which side of the firewall the wireless network belongs on. I tend to favor having an "ok to connect new object" button anyway, for security reasons...

UPnP is one of them evil Microsoft things :-) Is it the right one? What about security - how do you implement that correctly for this kind of device?

Nice idea, but what about this...

[pr0ntab]

Looks like a pen. Writes like a pen.

But it's not a pen. It contains 256 MB (or so) of flash, which is shared via bluetooth. A 10cm high gain antenna is hidden within the length of the pen itself, and powered by a single AAA battery. Walk by an enabled PC, optionally type in a password, and all your documents, your keyring, etc. are available.
Finally, as an added bonus, when you write on paper (or anything for that matter), you can choose to record your scribbles on the flash drive. Tiny gyroscopic sensors determine the motion of the pen across the page, and a pressure sensor determines whether the pen is against a writing surface. Each time you expose the ball point head it creates a new file, and when you retract it, it closes it. You can tell which file is which by the timestamps.

THAT would kick ass. And as embedded logic gets more powerful, you could have a personal web/email/jabber server running in there too.
A wireless iPod sounds nifty, but where's the innovation people?

Sounds like Roaming Profiles meet "next big thing"

[default luser]

Except you set it up once and can use it anywhere.

ANYWHERE.

Desktop, mobile, PDA, Cell phone, you name it. You're customized and ready to go, your OS and programs follow you ( The simpler ones run on the Xscale processor, the wireless link is just for shared data and I0 / screen updates, the more complex platforms get streamed boot data then program data, like your desktop )

We've been bitching at Intel for years for not making things portable enough, instead concentrating on bigger, hotter power hungry desktop chips.

Now Intel looks to be making a move to make everyone's standard data server run on a capable but not overkill Xscale.

Sure, there are some specialized things it can't process. Want to play a game? Mosey over to the desktop, your Xscale will stream the game files over in a few seconds, and while you play, fully configured with your key. Want to crunch a new movie into Divx or your favorite video codec? Your desktop will encode it and download it to the server in real time. Your desktop probably won't even have a hard disk, everything including the OS should be streamed from your handheld server.

* The key point here is this can be used as a seamless virtual session device ( ala Citrix ) plus fileserver plus desktop companion. Remote GUI sessions have always had problems catching on because the software is expensive and they require solid bandwidth. This has neither problem, and does more, so it's not dedicated hardware you're plunking down for. The battery life could be better, true, but for first generation >4 hours is impressive.

The main significance

[Lord_Dweomer]

The main significance that I see with a system such as this, where its use is merely to store data to be read/modified by other devices is that since now we can store this part of the computer, imagine how much power you could suddenly cram into a "handheld" device. Now I think we will see a small breakaway from that concept and into another concept which I feel will bridge the gap between handheld users and wearable users. Slowly we are seeing the modular wearable computer form. If all you need to hold in your hand is a screen/speakers/input device, suddenly you are working with significantly more space to play with if you remove the processor and HD and memory to a wireless unit stored in lets say.....a backpack.....or a purse....or a briefcase........or a jacket pocket. We can't always just move to smaller devices, sometimes we need to split them up, and then individually shrink the components, and this is a good step towards doing just that.

What if I don't want the thing in my pocket

[zorander]

Even at that size, It's annoying to add yet something else to my pockets. Three of my four pockets are currently consumed by a cell phone, wallet, and keys. This needs a better place to go. It's annoying to have technology in my pocket, especially if it is in any way fragile. Furthermore, does this thing generate any heat (which would make pocket use more uncomfortable).

Granted, most women have a bag of some sort with them, but men tend to go without. I'm not for any kind of subcataneous form factor, but I'd tolerate an anklet or something (don't ask me how to get hard drives into an anklet, just a thought). Even better, put it in my watch--I'm more likely to have that than my cell, and it's fairly well accepted to wear one.

Form factor must be considered. A Hard drive for your pocket isn't sexy, a normal enough looking watch that holds 80gigs whenever you're in 802.11/bluetooth range is--noone has to know.

Brian

A better mousetrap

[Anonymous Coward]

They're all missing the boat.

What we need are:
- Simple devices (a camera that just takes pictures, a MP3 player that just plays MP3's, a GPS that just spits out GPS data, a scanner that just scans, a printer that just prints)

- Cross-device scripting language, like "Javascript for Devices" or like Sun's old Java Bean Studio or even "Lego Mindstorms for Devices" or something like that

- Wireless, of course (short range)

- A PIN # to match all of my devices

- Rendezvous so that they all auto-recognize each other

This way, I could program my camera to talk to my GPS so that I know where that picture was taken, program my camera to talk to my cell phone so that when I double-click the shutter, email that picture to Mom. Program my printer to talk to my camera to print out pictures, program my car to email me when my odometer reaches time to change the oil, etc. etc.

When I go on vacation, I just strap my GPS and my portable hard drive to my belt and bring my camera, no more wondering where that picture was taken, no more worrying about storage.

Simple Devices and Cross-Device language.

Eliminate the PC and complexity of it. Why do I need to fire up my PC to print & save my pictures?

This company almost gets it. As does Intel, by the article.

DNET. You heard it here first.

Welcome to the next revolution.
williesleg@hotmail.com

I've worked out most of the details, don't have any prototypes yet, though. Soon.

Re:Not a "handheld killer"

[ocelotbob]

Exactly. Many of the people posting here just don't get it. This is about making things easier in the long run. No need to really worry about syncing your data; it's already been handled. Setup your mail program to export your iCal/vCal data every few minutes, and you've got your schedule everywhere. You don't have to remember to do it, because it's already been done. The only reason I said "stick it into your computer" is because of the charging issue. Probably a better phrasing would be stick it onto the charging cradle or the like. But now we're just picking nits here.

About the only gripe I have about the device is that it only uses 802.11b. Intel needs to step up to the plate and provide a device with 802.11a and g functionality as well; provide faster tranfer speeds so that one can stream higher quality media to their devices 11Mbps is a bit constraining when you're transferring media, but the 50Mbps+ a and g offer are better, or, like I feel will be the next thing in wireless, offer a way to use both a and g at the same time to get 100Mbps+ speeds out of wireless devices. Then one can really start to think about this device as a pervasive media server.

PDA perfection

[Ender_Stonebender]

As a portable computing junky, I've spent a lot of time thinking about what would make the "perfect PDA/Cell phone". This article has actually caused me to add a couple of things to it. Here's what I've come up with so far:

• Flip phone design similar to Kyocera's 7135 smartphone
• Total of 3 LCD screens:
  • Color LCD screen (inside) for main display, on the top half
  • Monochrome LCD that covers only an area big enough for handwriting recognition or virtual keyboard or status messages for games/integrated debuggers/whatever
  • Monochrome LCD screen along the top when closed, used for caller ID and alerts
• MP3 capability
• Some sort of wireless connection
• Integrated web server for entering data when there's a computer around, and a Java interface similar to VNCs for creating a virtual screen so that other programs can be used on other computers
• Capability to communicate with USB or serial for those who want it
• GPS-based alerts - these come two ways, "remind me to pick up my prescription when I'm near my drugstore" and "remind me about the meeting in time for me to get there"
• Massive amounts of storage for a PDA (for this type of device, I'd say 10 GB is "massive" enough)
• Two versions: one with an optical mouse on one side and buttons on the other and a built in thumb operated keyboard that uses a pointer to "tap" on things for those who don't like the stylus or can't/won't learn to deal with handwriting recognition, and one with a 4-way jog dial for scrolling through documents.
• Expandability - I feel that the CompactFlash+ format would be best (most room for additional electronics), but I haven't really looked deeply into it.

Of course, there are going to be some things I missed - feel free to point them out.

--Ender

Network While You're Networking ;)

[Taliesan999]

Makes for a whole bunch of interesting possibilites...

How long before someone develops a piece of software that lets you swap files with other such devices in the area automatically, maybe even search for specific files on other's shares?

Turn up to a lecture and just by being there you get a copy of the audio and notes streamed to your personal server.

Add an access mechanism to a television/home entertainment system. Boom, instant portable TIVO!

It's about time they caught up.

[linuxtelephony]

Unfortunately the site is gone now (I sold the site in 2000 and it got virtually no attention afterwards), but I wrote about technology like this back in 1999.

I hope they really come out with this technology for a reasonable price, and with a good modular design.

This type of device could be the central component of the mobile communications platform I was describing in a couple of articles.

Add a module for a wireless phone -- beauty here is that module could be CDMA, TDMA, GSM, pick the technology. Add another module for a pager if you need it. Yet another could be used for wireless data/internet.

Next, add your accessories -- color watch display, wireless headset for phone, handset for phone, wireless headphones, handheld (PDA style) display, digital camera (including video), small microphone for dictation (like some MP3 players have), keyboard, external extended battery pack (keeps internal battery charged, main power drawn from external pack until pack is drained), etc.

This "platform" could be marketed by just about everyone, from computer stores ala Fry's and CompUSA to electronic stores (Radio Shack) to wireless carriers.

Here's they key -- multitasking. This device would have to allow the use of the "phone" at the same time as the other modules such as camera (either still or video), PDA/watch, etc.

My wireless background obviously slants my opinion towards wireless phones etc, but I find it interesting how close things are getting to some of the ideas I've worked on or written about over the last 10 years. The potential for this is huge, if they deliver this right.

What's wrong with this picture?

[Animats]

Let me get this straight. I'm supposed to carry around this little box with all my data on it so that when I get somewhere that has a keyboard and screen and knows how to talk to this box, I can log in and access my data.

What's wrong with this picture? Why do I need to carry around this box? Why do I want to carry around data? That's what the Internet is for.

Remember Java-enabled jewelry with onboard crypto? The RSA "fob" ID device? Dallas Semiconductor buttons? Same functional capabilities, less to carry. All you really need is an ID device.

Ubiquitous computing looks more like "hurry up and find something that wastes compute power before we have to have another layoff". They need some better ideas over there.