This is news? The cygwin XF86 port has been around since long before May. Anyone who's installed cygwin this past year probably knows about it.
It didn't run rootless as of the last time I checked, which is quite a limitation. Has this changed? Hummingbird and Starnet don't have anything to worry about until then.
No one is going to flip through a thousand channels--though I'd argue a 1000-position knob is still more pleasant than up-down buttons--but what if you had a remote control with a color display and an iPod-like wheel, so you could zip through a list of channels and stop on one as its logo flashes past, and could instantly re-sort them alphabetically or grouped by content?
Why does it take half a second (or more) to transition from channel to channel when you want to skip past 50 of them? What if you could write the name of the channel you want to see by scribbling it on a pad--or scribbling it on the cellphone or PDA you're already holding? In the thirty seconds it took me to come up with those interfaces, I'm sure I've improved vastly on the current way of doing things. Up/down buttons and joypads for scrolling through slow, tedious menus aren't the only ways to skin the cat. What if you or I or some engineers worth their salt spent more than a minute thinking through this problem?
And the chronic VCR interface problem of "recording at a certain time" is an easy one to tackle--and the TiVo and other PVRs do largely tackle it--their interface problems lie elsewhere. All it took was rephrasing the problem as "recording a certain program".
Nobody wants to record "channel 74 from 9pm to 9:30pm". They want to record tomorrow night's "South Park". There's no reason in this day and age that a cheap TV attached to a VCR shouldn't be able to let me scribble or say aloud "record South Park", ask me which upcoming airings of South Park I want recorded via a handheld display like my PDA, and be done with it, shielding me from "start and end times", channel numbers, and zeroing the tape counter.
The Palm is an interesting example: it's a pretty good interface design for its kind, and apart from a Grafitti reference card, someone comfortable with a WIMP interface doesn't need an instruction manual in order to use it.
That explained a lot of its initial and ongoing success; as long as Palm buyers were businesspeople and others who use modern, mouse-oriented PCs regularly, the Palm was intuitive.
But it really isn't intuitive to a marginal PC user. What's intuitive about the "home" button and the "menu" button? What's obvious about clicking "Details.." to adjust the alarm setting for an appointment?
A paper phone book or weekly planner is intuitive, whereas a Palm simply does a good job of leveraging a user's already-learned skills in using a WIMP interface.
Instead of getting mad at your mom, find someone who will be patient and cheerful when showing her how to use the Palm's modified WIMP interface. It does take training, and reading a manual is only the best way to do that if a cheerful human teacher isn't available.
Don't you learn more easily when someone knowledgable and cheerful teaches you something than when you read a chapter of a book?
There's a weird, sad notion shared by many engineers--software, hardware, mechanical, electrical and otherwise--that the problem is with people, and that with time people will learn to think like engineers. 25 years into the personal computing era and the VCR era and the cable-TV era, it should be obvious that people are still people, and most will never be engineers. It's the engineering that has to adapt.
As I mentioned in another reply above (all the responses to my original post were so dismaying that I had to respond to all of them), Apple's iPod is complex and versatile, organizes its data more richly than any 200-CD changer, and yet has an interface consisting almost entirely of one knob and one button.
The Palm's interface is still the best PIM interface out there as far as usability goes--but it can still be made far, far more intuitive. It was a smart set of compromises when it was designed, making the best of cost constraints, technical limitations, and canny assumptions about its targeted customers at the time. Let me know when crossing out an item deletes it, like the Newton did years ago, or when flicking your finger across the page of an e-book turns the page. When interface moves like that are the norm rather than isolated frills atop a regular WIMP interface, nobody will need manuals. No general-purpose PIM should need a manual.
Ask why a CD changer needs "more buttons" than a single-disc CD player. What has changed in its mission? It plays music. Why can't it have an interface that isn't coupled so literally to its mechanical design? Shouldn't the interface instead be based on its purpose?
An intuitive 200-CD changer should be oriented to telling the user simply and straightforwardly what's currently playing, and asking her or him what they want to play next or which disc they want to eject.
Have a look at an iPod. It's an extraordinarily complex device: it can hold thousands of songs, organize them by artist, title and album, by how recently they've been played, and in other ways.
Count the buttons and knobs. It has fewer than a single-disc CD player or a 1974-vintage portable cassette player does. It's still not as easy as it could be, but it's still a superb lesson in how to design an inetrface based on how a device is used rather than how it's built.
I won't go into the distinction between assembling an early automobile and driving it.
But step back for a moment and think about what you said about a modern entertainment system. Does it need to be complex to use? What has changed from the days of the three-knob television or the five-button remote?
The only thing people do with a home theater setup 95% of the time is watch something or listen to something. There is absolutely no reason whatsoever that users should be forced to think in terms of the interactions between devices. Someone is either watching a bradcast, watching the thing on the tape in the VCR slot, or watching the thing in the DVD slot. Why is the channel metaphor broken when a user opts to watch the tape or the disc? Why is it necessary to "switch devices" and switch interfaces when going from VCR to DVD?
For a consumer product, if it needs a manual, the interface is too complex.
Back in the days before cordless phones and integrated answering machines, did you ever see a manual for a single-line telephone? Nope.
Before "home theater", did televisions need instructions beyond a card showing which antenna wire goes to which connector? Nope.
Have you ever rented a car and needed to read its owner's manual? Nope. What did it take to learn to drive a car? A few minutes' orientation and some practice on your own, right?
Except maybe for the one time three years ago when you cleaned your oven, have you ever felt the need to read the owners manual for it? Does your refrigerator need more instructions than the two sentences printed next to the temperature-setting knobs and the labels on the "fruit" and "meat" drawers?
How about an old Polaroid camera? An electric razor?
Sure, professional tools have always required training and big instruction books, whether it's a jet plane, a video editing console, a steamboat or Adobe Photoshop. But why do consumer e-mail and DVD players need more than a page of instructions?
The personal computer and the VCR trashed over 3,000 years of intuitive tool design. Before 1976, there was never a consumer product that needed a twenty-page instruction booklet (like a VCR), much less the shelf of books needed to operate a PC. Though it's understandable that something as complex as an office suite needs big manuals and user training, it's disgraceful that "wizard"-driven VCR programming has become a common feature in only the last five years, and appalling that anyone is expected to operate a typical home-theater setup. It's a wonder so many people manage to operate their main television these days. There's nothing intuitive or pleasant about the process.
If anyone with basic hand-eye coordination and an elemantary-school reading level can't operate a TiVo, then the answer is no. The TiVo's user interface isn't simple enough. With nearly all cable and satellite TV systems now transmitting listings on a side channel, not to mention dialup data transfer available to the TiVo itself, there is no reason at all that operation of a PVR should ever require a user to make use of ambiguous symbols on a one-way remote control or nonverbal cues onscreen. Anyone who has used a pocket calculator, a touch-tone phone or changed channels or adjusted volume on a TV should be able to use a TiVo without so much as touching a manual once the gadget has been connected to a TV and a power outlet.
It's not the users. It's the engineers. Get over it and do something about it, and stop using your technology background, whether self-taught over endless hours, or picked up in school, to demean and devalue others. People aren't going to "evolve" and be "trained" to spend hours and hours paging through manuals and fidgeting with lazily-designed tools; CompSci majors enjoy it as a sort of hobby. Other people have other hobbies, like playing outside. The purpose of a TiVo is to simplify the act of watching television, not to replace it with a puzzle game.
It's up to the devices to meet people at their own level, and any engineer who feels otherwise should find a new profession. Creating--and supporting--devices that people use is about helping people, not rubbing their faces in the dirt.
With a 20 MHz CPU and the expanded instruction set of the Z8 processor, it should also be able to spool hundreds of biorhythm charts out to my daisywheel in seconds and run Hunt the Wumpus really fast. This is nice!
I wonder if they're going to update CP/M to support all that RAM transparently. That would be sweet.
I think there's some missing the point going on here. From Sun's perspective (indeed, from a sysadmin's perspective), the lack of its own setup tools, relying on a command interface to change settings is a plus.
Metacity gives GNOME a chance to address one of its manageability flaws, the confilct between a desktop environment and the window manager. Which controls wallpaper? Screensavers? Why are there separate themes and theme settings interfaces for window chrome and the window contents?
It's because some power users high up in GNOME and window manager development--who usually aren't responsible for any machines beyond their own personal ones--like the flexibility of mixing and matching, and like pushing the bounds of what each component of their system can do. So overlapping--and conflicting--features get built.
This isn't the end of the world, but it does make a GNOME system more unwieldy than it has to be. KDE can run with several window managers, but it comes with one of its own that leaves configuration matters to KDE. GNOME hasn't had this yet. Enlightenment, sawmill and sawfish have been progressively better fits, but Sun and others who are moving to Metacity probably see it as a simpler route to getting a decent (GTK+ 2, anti-aliasing, multihead, accessibility-enabled) window manager seamlessly tied into GNOME than revamping Sawfish--and subsuming all of its configuration into GNOME--would be.
GNOME with Sawfish is a much tougher sell to a simplicity-minded CDE administrator than GNOME with Metacity will be, I suspect.
Good grief. How many copies of the GPL and LGPL are there for the reading on a typical, everyday Linux system? How many mentions of Stallman himself are tucked into the/usr/doc tree and accessible from command lines and the help menus of GUI programs? How many textfile histories of GNU and Linux? Dozens? Hundreds? That's not enough for his insatiable ego?
Maybe the community at large would be more receptive to Stallman's demands if he and the FSF crowd weren't so tone-deaf to good names for things. One of Stallman's less laudable contributions to Free Software is a tradition of terrible naming, one reflected even in heretical projects like KDE and Mozilla, and in more FSF-friendly matters like GNOME and GIMP. Renaming something ten years old to satisfy a Founding Father with ego issues is a tall enough order. Renaming it from something vaguely unappealing to something just awful-sounding is a bit much.
For all his monumental contributions, including his spearheading of this whole 100%-free Unix thing in the first place, maybe it's time to cut him loose. It's probably time for some kernel hackers to take some time off from Linux and finish up the HURD so the FSF can march off into a Hobbit-land where everyone does their word processing with TeX and GNU Emacs, and fiber-channel disk arrays are tools of fascism.
Then it'll be time to work on clean-room, GPL'ed drop-in replacements for all the GNU tools, call it LNG, or Linux is Not GNU, and be done with it.
I finish with a question. Since according to Stallman, FSF computers can't have Bitkeeper on them because it's non-Free software (a point well made and an admirable stance--as always it's the histrionic Stalinist demand that everyone else hobble themselves in deference to that point that grates), how do they work at all? Do they run systems with a Free, open-source BIOS and Free, open-source firmware on the video cards that drive their (no doubt) text-only displays too?
I'd be cranky too if I worked all day in Emacs on a daisywheel printing TTY.
Three years later, I'm still mystified by the attention Abiword gets. It even gets press coverage.
It's not even a word processor by late-1980s standards. No table support! No floating footnotes! The column support doesn't seem to allow changing the number of columns midstream--it's all or nothing.
No merge functionality! (Oh, but there are two optional, unbundled scripting plugins you can theoretically write your own merge function with--except that there's no user-defined field support, either, so any merge fields in a document would be ad-hoc, unprotected, and would show up as spelling errors.)
Great, so it's "lightweight" and starts up quickly, and it's cross-platform. Yipee. But I remember in 1988 it was pretty fair to expect a graphical word processor--even on the Amiga and the C64--to support tables and footnotes, mail merging and real, multiple-layouts-per-page column support.
Don't get me wrong. It's nice of the Abiword team to put their time into writing software they obviously find and useful, and it is nice to see a solid, genuinely useful embeddable GTK+ richtext widget come out of this, but can we please stop mentioning it in the same breath as word processors?
One thing the guy didn't mention (unless my eyes are going) was the specs of the scanner. If it's a low-end (say, 300 or 600 dpi) scanner, I'm curious as to whether higher-density scanners have higher-resolution CCDs. It's a terrible point-and shoot, but large-format photographers would be very ineterested in, say, an affordable 4000x4000 or even higher-resolution camera like this, twenty-second exposure times and all. It would be a terrific gadget for landscapes, architectural photos, and still-life studio work. At the current 2000x2000, of course, it's just a curiosity.
Re:Still unclear on why tape recorders are no good
on
USB Audio Recorders?
·
· Score: 3, Interesting
1. Anyone handy with a knife and a screwdriver--and I mean anyone--can build an AM or shortwave transmitter with miles of range from a few dollars' worth of parts, and can do so for even less with salvaged parts. FM's not forbiddingly complex, but it does need bigger, more precisely measured antennas and bigger transformers, and a lot more juice.
2. Shortwave receivers are expensive in the United States. The receivers used in countries like India--espeically in remote parts of the country that aren't even served by FM--are cheap. Rural India is not the same as suburban Bangalore. As with western China, much of Siberia and much of Africa, where the population is either sparse or very poor, the short range, expensive equipment, and high power requirements of FM don't make much sense.
3. Who said anything about reel-to-reel tape? Who said anything about "professional-quality" audio? This is a community-radio project. They're broadcasting news, weather, and farming reports. For half a century, the world was more than happy with the sound quality of AM broadcasts. People even listened to music on it. Billions of people still do. Plain old consumer-grade cassette tape should be more than adequate for something like this. Minidisc media are expensive. The recorders and playback devices are expensive both to obtain and to repair. And they need repair a whole lot more often than a $20 cassette recorder. Not to mention training involved, and the relative vandalism and theft risks.
Radio journalists worked for many decades with plain cassette recorders, without even Dolby B. I think Indian villagers can, too, especially for broadcasting speech.
Still unclear on why tape recorders are no good
on
USB Audio Recorders?
·
· Score: 2
I'm still not clear where digital audio comes into this. Is this data requested for broadcast being read aloud and digitized by a central team and being e-mailed out to the villages? It sounded more like text was being sent to the villages.
Even so, I still don't see why faster-than-realtime dubbing is at all important to something like this, or why unreliable, developing-world internet connectivity (read: slow modems doing UUCP or such) is seen as a good way to move audio around, even 8 kbit/sec low-quality audio. Wouldn't broadcasting the audio via ham radio on a regular schedule be much cheaper and more reliable? To fetch the appropriate audio for a given village in order to rebroadcast it, all you'd need is a cheap tape recorder, a cheap shortwave receiver with BFO, and.. well.. a lamp timer to start recording at that village's assigned time. No phone lines, no computers deployed in villages, and no $300 minidisc recorders that'll be broken within six months.
Need the ability to edit in locally-produced audio, too? Okay: give them a second cassette recorder, a couple of blank tapes and a dubbing cable.
This would cost a fraction as much as the computer/minidisk arrangement, be more reliable, more maintainable and more sustainable.
Less glamorous, I'll grant, and maybe it doesn't give IT people involved in the aid organization much to do, but imagine the warm fuzzy feeling you'll get when followup site inspections a few years down the road find equipment still in working order.
Am I missing some key thing that necessitates computers and digital media in each village?
Good GOD, why does this need high-tech?
on
USB Audio Recorders?
·
· Score: 3, Insightful
Why does a local radio station need computer connectivity and minidisks? What on earth is wrong with a cheap AM or shortwave transmitter? And what's wrong with analog audio cassettes? Apart from the power source for the transmitter (and AM and shortwave draw a whole lot less power than FM), this equipment should cost maybe $50 per station, including an old microphone and 2-channel mixing board. And the equipment for such a station can be fixed by anyone, using salvaged parts.
And as for using it to read internet content, how is this village connecting to the internet? If they've got reasonable phones, I guess you could do ftp-by-mail and fetch it overnight by UUCP on a 386. Surely realtime web surfing is an expensive pipe dream in these places. Why on earth is a serious computer with USB at all necessary? In a village where a modern (say, Pentium) computer could be put to much much better use, like public e-mail, research, weather and agricultural data analysis and so forth, setting up a needlessly fancy computerized local radio station is idiocy.
What part of reading news and information that (in a remote village in the developing world) will almost certainly come delayed in the form of text email requires a computer hooked to the radio station?
Can't someone just print stuff out from the village computer (on a reliable, unglamorous, and cheap-to-supply-with-ink dot-matrix printer) and read it on the air on an AM/shortwave station with ten times the range running on a tenth the power of FM? And if they've got to record it for rebroadcast or something, what's wrong with a cheap mono cassette boombox from a backpacker's duffel bag?
There is also credible evidence that Basque whaling fleets were regularly working the waters off Newfoundland by the late 1300s. Though it didn't have much of a direct impact on North America, it would have to be considered one of the factors that allowed the Spanish to get to the Americas at all a century later, relying as they did on Basque sailors and navigators who, through this lens, would have already crossed the Atlantic.
Whalers were not known for sharing information on where they fished. It was important and closely-guarded proprietary information.
This strategy of seeding the tsetse fly population with sterile flies in order to reduce successful reproduction is clever, but it would be much more fun if they released cane toads to eat the flies instead.
A hundred or so shipped in from Australia should do the trick.
StarOffice/OpenOffice 6 Macro and Applications Programming (Possibly separate books on doing so with Java, Javascript, and their VB clone).
Setting up secure home and small-business wireless networks (Running VPNs and the like over 802.11-series networks, covering Windows, Unix and MacOS X approaches, and both hardware and software solutions. commercial and free)
It was good for Mozilla and Netscape 6.x when jwz left. It was under him that the project scope kept changing, the notion of scheduled milestone releases went out the window, and most of the good longtime coders left.
It was after he left that the team began releasing frequent milestone builds, stopped adding major new features to the project plan, and.. showed signs of having a plan.
The Mozilla/Netscape 6 project is still a mess, with bug fixes and addition of missing features slated for a given milestone pushed off to infinity on a regular basis. But without jwz, it at least resembles a project and has produced what is now a decent browser and mail/news client.
Mr. Zawinski is now running a bar, and the world of software development is blissfully free of his project management "skills".
Alan Cox--who unlike jwz is a really sharp coder and a good project leader--is showing himself to be just as much a child, spoiled and twisted by too much time spent in academic computing, shooting his mouth off before he's got a real situation to evaluate. Hey. If AOL turns Red Hat into an unpleasant place by changing its focus in distressing ways, or by engaging in massive, traumatic waves of layoffs, of course he'd be right in leaving. If Red Hat lets him pick his projects and AOL instead wants him to port the AIM stock ticker to KDE or sit in meetings all day, of course he'd be justified in leaving.
But this knee-jerk aversion to a parent company just because it's a big company? Or because of AOL's commitment to actual ease of use that Cox, jwz and RMS all abhor?
What if AOL is trying to assemble all the pieces necessary to go after Microsoft with Free Software? Doesn't Red Hat also employ some Postgres maintainers? If they bought Staroffice/Openoffice from Sun, they'd be on their way to something mighty compelling. If an AOL-owned Red Hat lets him continue working on low-level kernel pieces and device drivers while they fund an aggressive desktop-oriented Red Hat, why wouldn't he want to come along for the ride? Because they also own an old-line record label and film studio with rabidly protected intellectual property? Okay.
I wish the best of luck to any company, school or organization that wants these guys on its payroll.
Last time I checked, NCD, Wyse, Tektronix and IBM all sold combo X/ICA terminals. X for the Linux, ICA for the Citrix stuff and, increasingly, direct support for Windows Terminal Server.ther than Sun
I'd be surprised if anyone in the hardware X terminal business other than Sun didn't also support at least ICA at this point.
Another option everyone has forgotten these days is Tarantella, the SCO product, which runs as a server both under Unix (including Linux) and on Windows and can serve either kind of application via RDP and X11 if I recall.
It's a buffer overflow in affecting the 9i Application Server--specifically, a PL/SQL Apache module--and not the database. Still a Bad Thing, but not the same thing.
The crack regarding "still waiting for [your] Network Computer" is pretty dopey. Ellison's NIC Company has been shipping them going on two years now.
You'd think they'd be a big hit with the Slashdot set seeing as they boot Linux with X off a CD, and have Ethernet, USB, a modem and VGA support built in, all for $200. I guess lame jokes predicated on them not existing are more fun.
These are teensy little flash memory cards with USB connectors and IDE-over-USB emulation like most of today's flash-memory technologies.
The prices are the same or a smidgen higher than the same size CompactFlash, Smartmedia, Memory Stick or MMC cards.
And they've been out for more than a year, though the 512MB and 1GB models are pretty recent. The idea is they're an alternative to shuttling a small batch of files around on a Zip disk or such, or burning a CD.
As for actual hard drives, for half that $900 figure you can get a PC Card drive for your laptop that holds 5GB though like IBM Microdrives it's obviously a bit more delicate. And you can get pocket-sized 30GB Firewire and USB 2 drives for the same $400 or so these days.
What doesn't get posted to Slashdot these days? When will we be hearing about someone discovering Dim Sum? Or asking for resources on learning how to drive a stick-shift?
Damn straight, re: the ubiquity of commercial iCalendar servers. But as for knocking Ximian for not filliong the GPL iCalendar server gap:
Apparently Ximian is trying to focus on one thing at a time. Right now it's the notion of a Linux/Unix desktop that can work in current Windows/Mac environments, which means a desktop interface and essential apps like groupware, spreadsheets and so forth.
After that reaches a certain level of quality (GNOME 2.0 and perhaps a second major release of Evolution), their next announced project is an implementation of.NET, which besides being a server technology also happens to be a technology key to ongoing desktop interoperability in heterogeneous environments.
Ximian isn't in the server software business. Granted, I could make some snarky comments about their interface design chops, but I won't knock them for having a focus.
For a bunch of techies, there's not much imagination here. I can't get to the information on the Sony unit in question, but there are plenty of uses for a TCP/IP-enabled camera or camcorder.
Ricoh already sells a digital still camera with an IP stack, mail and FTP clients and a PC Card slot that will take a netowrk or modem card. Gimmicky? Nope. For surfing and checking your mail? Nope. What it lets you do is email or FTP images directly from the camera. If you're a traveling real-estate or newspaper photographer, what could be more immediate than that? No trip back to the hotel, no stopping at a service bureau--just connect to your data-enabled cellphone (or relay via Bluetooth! or a nearby 802.11b access point!) and send out that lo-res picture you just took.
Just because the US is still seemingly years away from 2.5G or 3G wireless data service, that doesn't mean Japan is. Hell, if Metricom hadn't called it a day, even that would be a compelling way to send out brief low-res clips.
Sure, we're a ways away from being able to transmit full-resolution DV over the net from a handheld camera, but that doesn't mean putting the functionality in now won't find an audience for the things it can do now.
Those videophones TV war reporters have been using during the past month in Afghanistan aren't hi-res or high-bandwidth either, but that didn't make them useless.
No, not if it's existed for decades. It's what's referred to as a "mainframe". You know. An expandable number of processor boards running under an operating system that can treat them as any number of single-processor or multiprocessor machines, with the ability to reassign processes between CPUs.
The Unix world has had them for a long time, too. Modern examples include Sun's higher-end servers, which support hot-swappable and hot-pluggable processors and memory.
Doing it with x86 processors and standard x86 OSes like x86 Unixes and Windows is less common but I believe Compaq and maybe Unisys can sell you machines that can do it, too, with one or several instances of the OS running at once.
This hatdware approach is not quite the same as VMWare's server products, which do it via software and don't limit you to one OS per processor or block of processors. It in turn mimics other decades-old mainframe operating environments in approach.
Slashdot Mirror
This is news? The cygwin XF86 port has been around since long before May. Anyone who's installed cygwin this past year probably knows about it.
It didn't run rootless as of the last time I checked, which is quite a limitation. Has this changed? Hummingbird and Starnet don't have anything to worry about until then.
No one is going to flip through a thousand channels--though I'd argue a 1000-position knob is still more pleasant than up-down buttons--but what if you had a remote control with a color display and an iPod-like wheel, so you could zip through a list of channels and stop on one as its logo flashes past, and could instantly re-sort them alphabetically or grouped by content?
Why does it take half a second (or more) to transition from channel to channel when you want to skip past 50 of them? What if you could write the name of the channel you want to see by scribbling it on a pad--or scribbling it on the cellphone or PDA you're already holding? In the thirty seconds it took me to come up with those interfaces, I'm sure I've improved vastly on the current way of doing things. Up/down buttons and joypads for scrolling through slow, tedious menus aren't the only ways to skin the cat. What if you or I or some engineers worth their salt spent more than a minute thinking through this problem?
And the chronic VCR interface problem of "recording at a certain time" is an easy one to tackle--and the TiVo and other PVRs do largely tackle it--their interface problems lie elsewhere. All it took was rephrasing the problem as "recording a certain program".
Nobody wants to record "channel 74 from 9pm to 9:30pm". They want to record tomorrow night's "South Park". There's no reason in this day and age that a cheap TV attached to a VCR shouldn't be able to let me scribble or say aloud "record South Park", ask me which upcoming airings of South Park I want recorded via a handheld display like my PDA, and be done with it, shielding me from "start and end times", channel numbers, and zeroing the tape counter.
The Palm is an interesting example: it's a pretty good interface design for its kind, and apart from a Grafitti reference card, someone comfortable with a WIMP interface doesn't need an instruction manual in order to use it.
That explained a lot of its initial and ongoing success; as long as Palm buyers were businesspeople and others who use modern, mouse-oriented PCs regularly, the Palm was intuitive.
But it really isn't intuitive to a marginal PC user. What's intuitive about the "home" button and the "menu" button? What's obvious about clicking "Details.." to adjust the alarm setting for an appointment?
A paper phone book or weekly planner is intuitive, whereas a Palm simply does a good job of leveraging a user's already-learned skills in using a WIMP interface.
Instead of getting mad at your mom, find someone who will be patient and cheerful when showing her how to use the Palm's modified WIMP interface. It does take training, and reading a manual is only the best way to do that if a cheerful human teacher isn't available.
Don't you learn more easily when someone knowledgable and cheerful teaches you something than when you read a chapter of a book?
There's a weird, sad notion shared by many engineers--software, hardware, mechanical, electrical and otherwise--that the problem is with people, and that with time people will learn to think like engineers. 25 years into the personal computing era and the VCR era and the cable-TV era, it should be obvious that people are still people, and most will never be engineers. It's the engineering that has to adapt.
As I mentioned in another reply above (all the responses to my original post were so dismaying that I had to respond to all of them), Apple's iPod is complex and versatile, organizes its data more richly than any 200-CD changer, and yet has an interface consisting almost entirely of one knob and one button.
The Palm's interface is still the best PIM interface out there as far as usability goes--but it can still be made far, far more intuitive. It was a smart set of compromises when it was designed, making the best of cost constraints, technical limitations, and canny assumptions about its targeted customers at the time. Let me know when crossing out an item deletes it, like the Newton did years ago, or when flicking your finger across the page of an e-book turns the page. When interface moves like that are the norm rather than isolated frills atop a regular WIMP interface, nobody will need manuals. No general-purpose PIM should need a manual.
Ask why a CD changer needs "more buttons" than a single-disc CD player. What has changed in its mission? It plays music. Why can't it have an interface that isn't coupled so literally to its mechanical design? Shouldn't the interface instead be based on its purpose?
An intuitive 200-CD changer should be oriented to telling the user simply and straightforwardly what's currently playing, and asking her or him what they want to play next or which disc they want to eject.
Have a look at an iPod. It's an extraordinarily complex device: it can hold thousands of songs, organize them by artist, title and album, by how recently they've been played, and in other ways.
Count the buttons and knobs. It has fewer than a single-disc CD player or a 1974-vintage portable cassette player does. It's still not as easy as it could be, but it's still a superb lesson in how to design an inetrface based on how a device is used rather than how it's built.
I won't go into the distinction between assembling an early automobile and driving it.
But step back for a moment and think about what you said about a modern entertainment system. Does it need to be complex to use? What has changed from the days of the three-knob television or the five-button remote?
The only thing people do with a home theater setup 95% of the time is watch something or listen to something. There is absolutely no reason whatsoever that users should be forced to think in terms of the interactions between devices. Someone is either watching a bradcast, watching the thing on the tape in the VCR slot, or watching the thing in the DVD slot. Why is the channel metaphor broken when a user opts to watch the tape or the disc? Why is it necessary to "switch devices" and switch interfaces when going from VCR to DVD?
No, it's a design failure, and it's shameful.
For a consumer product, if it needs a manual, the interface is too complex.
Back in the days before cordless phones and integrated answering machines, did you ever see a manual for a single-line telephone? Nope.
Before "home theater", did televisions need instructions beyond a card showing which antenna wire goes to which connector? Nope.
Have you ever rented a car and needed to read its owner's manual? Nope. What did it take to learn to drive a car? A few minutes' orientation and some practice on your own, right?
Except maybe for the one time three years ago when you cleaned your oven, have you ever felt the need to read the owners manual for it? Does your refrigerator need more instructions than the two sentences printed next to the temperature-setting knobs and the labels on the "fruit" and "meat" drawers?
How about an old Polaroid camera? An electric razor?
Sure, professional tools have always required training and big instruction books, whether it's a jet plane, a video editing console, a steamboat or Adobe Photoshop. But why do consumer e-mail and DVD players need more than a page of instructions?
The personal computer and the VCR trashed over 3,000 years of intuitive tool design. Before 1976, there was never a consumer product that needed a twenty-page instruction booklet (like a VCR), much less the shelf of books needed to operate a PC. Though it's understandable that something as complex as an office suite needs big manuals and user training, it's disgraceful that "wizard"-driven VCR programming has become a common feature in only the last five years, and appalling that anyone is expected to operate a typical home-theater setup. It's a wonder so many people manage to operate their main television these days. There's nothing intuitive or pleasant about the process.
If anyone with basic hand-eye coordination and an elemantary-school reading level can't operate a TiVo, then the answer is no. The TiVo's user interface isn't simple enough. With nearly all cable and satellite TV systems now transmitting listings on a side channel, not to mention dialup data transfer available to the TiVo itself, there is no reason at all that operation of a PVR should ever require a user to make use of ambiguous symbols on a one-way remote control or nonverbal cues onscreen. Anyone who has used a pocket calculator, a touch-tone phone or changed channels or adjusted volume on a TV should be able to use a TiVo without so much as touching a manual once the gadget has been connected to a TV and a power outlet.
It's not the users. It's the engineers. Get over it and do something about it, and stop using your technology background, whether self-taught over endless hours, or picked up in school, to demean and devalue others. People aren't going to "evolve" and be "trained" to spend hours and hours paging through manuals and fidgeting with lazily-designed tools; CompSci majors enjoy it as a sort of hobby. Other people have other hobbies, like playing outside. The purpose of a TiVo is to simplify the act of watching television, not to replace it with a puzzle game.
It's up to the devices to meet people at their own level, and any engineer who feels otherwise should find a new profession. Creating--and supporting--devices that people use is about helping people, not rubbing their faces in the dirt.
With a 20 MHz CPU and the expanded instruction set of the Z8 processor, it should also be able to spool hundreds of biorhythm charts out to my daisywheel in seconds and run Hunt the Wumpus really fast. This is nice!
I wonder if they're going to update CP/M to support all that RAM transparently. That would be sweet.
I think there's some missing the point going on here. From Sun's perspective (indeed, from a sysadmin's perspective), the lack of its own setup tools, relying on a command interface to change settings is a plus.
Metacity gives GNOME a chance to address one of its manageability flaws, the confilct between a desktop environment and the window manager. Which controls wallpaper? Screensavers? Why are there separate themes and theme settings interfaces for window chrome and the window contents?
It's because some power users high up in GNOME and window manager development--who usually aren't responsible for any machines beyond their own personal ones--like the flexibility of mixing and matching, and like pushing the bounds of what each component of their system can do. So overlapping--and conflicting--features get built.
This isn't the end of the world, but it does make a GNOME system more unwieldy than it has to be. KDE can run with several window managers, but it comes with one of its own that leaves configuration matters to KDE. GNOME hasn't had this yet. Enlightenment, sawmill and sawfish have been progressively better fits, but Sun and others who are moving to Metacity probably see it as a simpler route to getting a decent (GTK+ 2, anti-aliasing, multihead, accessibility-enabled) window manager seamlessly tied into GNOME than revamping Sawfish--and subsuming all of its configuration into GNOME--would be.
GNOME with Sawfish is a much tougher sell to a simplicity-minded CDE administrator than GNOME with Metacity will be, I suspect.
Good grief. How many copies of the GPL and LGPL are there for the reading on a typical, everyday Linux system? How many mentions of Stallman himself are tucked into the /usr/doc tree and accessible from command lines and the help menus of GUI programs? How many textfile histories of GNU and Linux? Dozens? Hundreds? That's not enough for his insatiable ego?
Maybe the community at large would be more receptive to Stallman's demands if he and the FSF crowd weren't so tone-deaf to good names for things. One of Stallman's less laudable contributions to Free Software is a tradition of terrible naming, one reflected even in heretical projects like KDE and Mozilla, and in more FSF-friendly matters like GNOME and GIMP. Renaming something ten years old to satisfy a Founding Father with ego issues is a tall enough order. Renaming it from something vaguely unappealing to something just awful-sounding is a bit much.
For all his monumental contributions, including his spearheading of this whole 100%-free Unix thing in the first place, maybe it's time to cut him loose. It's probably time for some kernel hackers to take some time off from Linux and finish up the HURD so the FSF can march off into a Hobbit-land where everyone does their word processing with TeX and GNU Emacs, and fiber-channel disk arrays are tools of fascism.
Then it'll be time to work on clean-room, GPL'ed drop-in replacements for all the GNU tools, call it LNG, or Linux is Not GNU, and be done with it.
I finish with a question. Since according to Stallman, FSF computers can't have Bitkeeper on them because it's non-Free software (a point well made and an admirable stance--as always it's the histrionic Stalinist demand that everyone else hobble themselves in deference to that point that grates), how do they work at all? Do they run systems with a Free, open-source BIOS and Free, open-source firmware on the video cards that drive their (no doubt) text-only displays too?
I'd be cranky too if I worked all day in Emacs on a daisywheel printing TTY.
Alas, if you read that manual section on creating "tables", it turns out not to be table functionality at all.
It's just telling users how to make very basic table layouts by using tabs and the overline/underline styles.
Kind of like doing tables on a TRS-80 in 1982.
Three years later, I'm still mystified by the attention Abiword gets. It even gets press coverage.
It's not even a word processor by late-1980s standards. No table support! No floating footnotes! The column support doesn't seem to allow changing the number of columns midstream--it's all or nothing.
No merge functionality! (Oh, but there are two optional, unbundled scripting plugins you can theoretically write your own merge function with--except that there's no user-defined field support, either, so any merge fields in a document would be ad-hoc, unprotected, and would show up as spelling errors.)
Great, so it's "lightweight" and starts up quickly, and it's cross-platform. Yipee. But I remember in 1988 it was pretty fair to expect a graphical word processor--even on the Amiga and the C64--to support tables and footnotes, mail merging and real, multiple-layouts-per-page column support.
Don't get me wrong. It's nice of the Abiword team to put their time into writing software they obviously find and useful, and it is nice to see a solid, genuinely useful embeddable GTK+ richtext widget come out of this, but can we please stop mentioning it in the same breath as word processors?
One thing the guy didn't mention (unless my eyes are going) was the specs of the scanner. If it's a low-end (say, 300 or 600 dpi) scanner, I'm curious as to whether higher-density scanners have higher-resolution CCDs. It's a terrible point-and shoot, but large-format photographers would be very ineterested in, say, an affordable 4000x4000 or even higher-resolution camera like this, twenty-second exposure times and all. It would be a terrific gadget for landscapes, architectural photos, and still-life studio work. At the current 2000x2000, of course, it's just a curiosity.
1. Anyone handy with a knife and a screwdriver--and I mean anyone--can build an AM or shortwave transmitter with miles of range from a few dollars' worth of parts, and can do so for even less with salvaged parts. FM's not forbiddingly complex, but it does need bigger, more precisely measured antennas and bigger transformers, and a lot more juice.
2. Shortwave receivers are expensive in the United States. The receivers used in countries like India--espeically in remote parts of the country that aren't even served by FM--are cheap. Rural India is not the same as suburban Bangalore. As with western China, much of Siberia and much of Africa, where the population is either sparse or very poor, the short range, expensive equipment, and high power requirements of FM don't make much sense.
3. Who said anything about reel-to-reel tape? Who said anything about "professional-quality" audio? This is a community-radio project. They're broadcasting news, weather, and farming reports. For half a century, the world was more than happy with the sound quality of AM broadcasts. People even listened to music on it. Billions of people still do. Plain old consumer-grade cassette tape should be more than adequate for something like this. Minidisc media are expensive. The recorders and playback devices are expensive both to obtain and to repair. And they need repair a whole lot more often than a $20 cassette recorder. Not to mention training involved, and the relative vandalism and theft risks.
Radio journalists worked for many decades with plain cassette recorders, without even Dolby B. I think Indian villagers can, too, especially for broadcasting speech.
I'm still not clear where digital audio comes into this. Is this data requested for broadcast being read aloud and digitized by a central team and being e-mailed out to the villages? It sounded more like text was being sent to the villages.
Even so, I still don't see why faster-than-realtime dubbing is at all important to something like this, or why unreliable, developing-world internet connectivity (read: slow modems doing UUCP or such) is seen as a good way to move audio around, even 8 kbit/sec low-quality audio. Wouldn't broadcasting the audio via ham radio on a regular schedule be much cheaper and more reliable? To fetch the appropriate audio for a given village in order to rebroadcast it, all you'd need is a cheap tape recorder, a cheap shortwave receiver with BFO, and.. well.. a lamp timer to start recording at that village's assigned time. No phone lines, no computers deployed in villages, and no $300 minidisc recorders that'll be broken within six months.
Need the ability to edit in locally-produced audio, too? Okay: give them a second cassette recorder, a couple of blank tapes and a dubbing cable.
This would cost a fraction as much as the computer/minidisk arrangement, be more reliable, more maintainable and more sustainable.
Less glamorous, I'll grant, and maybe it doesn't give IT people involved in the aid organization much to do, but imagine the warm fuzzy feeling you'll get when followup site inspections a few years down the road find equipment still in working order.
Am I missing some key thing that necessitates computers and digital media in each village?
Why does a local radio station need computer connectivity and minidisks? What on earth is wrong with a cheap AM or shortwave transmitter? And what's wrong with analog audio cassettes? Apart from the power source for the transmitter (and AM and shortwave draw a whole lot less power than FM), this equipment should cost maybe $50 per station, including an old microphone and 2-channel mixing board. And the equipment for such a station can be fixed by anyone, using salvaged parts.
And as for using it to read internet content, how is this village connecting to the internet? If they've got reasonable phones, I guess you could do ftp-by-mail and fetch it overnight by UUCP on a 386. Surely realtime web surfing is an expensive pipe dream in these places. Why on earth is a serious computer with USB at all necessary? In a village where a modern (say, Pentium) computer could be put to much much better use, like public e-mail, research, weather and agricultural data analysis and so forth, setting up a needlessly fancy computerized local radio station is idiocy.
What part of reading news and information that (in a remote village in the developing world) will almost certainly come delayed in the form of text email requires a computer hooked to the radio station?
Can't someone just print stuff out from the village computer (on a reliable, unglamorous, and cheap-to-supply-with-ink dot-matrix printer) and read it on the air on an AM/shortwave station with ten times the range running on a tenth the power of FM? And if they've got to record it for rebroadcast or something, what's wrong with a cheap mono cassette boombox from a backpacker's duffel bag?
My god, have aid agencies gone insane?
There is also credible evidence that Basque whaling fleets were regularly working the waters off Newfoundland by the late 1300s. Though it didn't have much of a direct impact on North America, it would have to be considered one of the factors that allowed the Spanish to get to the Americas at all a century later, relying as they did on Basque sailors and navigators who, through this lens, would have already crossed the Atlantic.
Whalers were not known for sharing information on where they fished. It was important and closely-guarded proprietary information.
This strategy of seeding the tsetse fly population with sterile flies in order to reduce successful reproduction is clever, but it would be much more fun if they released cane toads to eat the flies instead.
A hundred or so shipped in from Australia should do the trick.
It was good for Mozilla and Netscape 6.x when jwz left. It was under him that the project scope kept changing, the notion of scheduled milestone releases went out the window, and most of the good longtime coders left.
It was after he left that the team began releasing frequent milestone builds, stopped adding major new features to the project plan, and.. showed signs of having a plan.
The Mozilla/Netscape 6 project is still a mess, with bug fixes and addition of missing features slated for a given milestone pushed off to infinity on a regular basis. But without jwz, it at least resembles a project and has produced what is now a decent browser and mail/news client.
Mr. Zawinski is now running a bar, and the world of software development is blissfully free of his project management "skills".
Alan Cox--who unlike jwz is a really sharp coder and a good project leader--is showing himself to be just as much a child, spoiled and twisted by too much time spent in academic computing, shooting his mouth off before he's got a real situation to evaluate. Hey. If AOL turns Red Hat into an unpleasant place by changing its focus in distressing ways, or by engaging in massive, traumatic waves of layoffs, of course he'd be right in leaving. If Red Hat lets him pick his projects and AOL instead wants him to port the AIM stock ticker to KDE or sit in meetings all day, of course he'd be justified in leaving.
But this knee-jerk aversion to a parent company just because it's a big company? Or because of AOL's commitment to actual ease of use that Cox, jwz and RMS all abhor?
What if AOL is trying to assemble all the pieces necessary to go after Microsoft with Free Software? Doesn't Red Hat also employ some Postgres maintainers? If they bought Staroffice/Openoffice from Sun, they'd be on their way to something mighty compelling. If an AOL-owned Red Hat lets him continue working on low-level kernel pieces and device drivers while they fund an aggressive desktop-oriented Red Hat, why wouldn't he want to come along for the ride? Because they also own an old-line record label and film studio with rabidly protected intellectual property? Okay.
I wish the best of luck to any company, school or organization that wants these guys on its payroll.
Last time I checked, NCD, Wyse, Tektronix and IBM all sold combo X/ICA terminals. X for the Linux, ICA for the Citrix stuff and, increasingly, direct support for Windows Terminal Server.ther than Sun
I'd be surprised if anyone in the hardware X terminal business other than Sun didn't also support at least ICA at this point.
Another option everyone has forgotten these days is Tarantella, the SCO product, which runs as a server both under Unix (including Linux) and on Windows and can serve either kind of application via RDP and X11 if I recall.
You'd think they'd be a big hit with the Slashdot set seeing as they boot Linux with X off a CD, and have Ethernet, USB, a modem and VGA support built in, all for $200. I guess lame jokes predicated on them not existing are more fun.
These are teensy little flash memory cards with USB connectors and IDE-over-USB emulation like most of today's flash-memory technologies.
The prices are the same or a smidgen higher than the same size CompactFlash, Smartmedia, Memory Stick or MMC cards.
And they've been out for more than a year, though the 512MB and 1GB models are pretty recent. The idea is they're an alternative to shuttling a small batch of files around on a Zip disk or such, or burning a CD.
As for actual hard drives, for half that $900 figure you can get a PC Card drive for your laptop that holds 5GB though like IBM Microdrives it's obviously a bit more delicate. And you can get pocket-sized 30GB Firewire and USB 2 drives for the same $400 or so these days.
What doesn't get posted to Slashdot these days? When will we be hearing about someone discovering Dim Sum? Or asking for resources on learning how to drive a stick-shift?
Damn straight, re: the ubiquity of commercial iCalendar servers. But as for knocking Ximian for not filliong the GPL iCalendar server gap:
.NET, which besides being a server technology also happens to be a technology key to ongoing desktop interoperability in heterogeneous environments.
Apparently Ximian is trying to focus on one thing at a time. Right now it's the notion of a Linux/Unix desktop that can work in current Windows/Mac environments, which means a desktop interface and essential apps like groupware, spreadsheets and so forth.
After that reaches a certain level of quality (GNOME 2.0 and perhaps a second major release of Evolution), their next announced project is an implementation of
Ximian isn't in the server software business. Granted, I could make some snarky comments about their interface design chops, but I won't knock them for having a focus.
For a bunch of techies, there's not much imagination here. I can't get to the information on the Sony unit in question, but there are plenty of uses for a TCP/IP-enabled camera or camcorder.
Ricoh already sells a digital still camera with an IP stack, mail and FTP clients and a PC Card slot that will take a netowrk or modem card. Gimmicky? Nope. For surfing and checking your mail? Nope. What it lets you do is email or FTP images directly from the camera. If you're a traveling real-estate or newspaper photographer, what could be more immediate than that? No trip back to the hotel, no stopping at a service bureau--just connect to your data-enabled cellphone (or relay via Bluetooth! or a nearby 802.11b access point!) and send out that lo-res picture you just took.
Just because the US is still seemingly years away from 2.5G or 3G wireless data service, that doesn't mean Japan is. Hell, if Metricom hadn't called it a day, even that would be a compelling way to send out brief low-res clips.
Sure, we're a ways away from being able to transmit full-resolution DV over the net from a handheld camera, but that doesn't mean putting the functionality in now won't find an audience for the things it can do now.
Those videophones TV war reporters have been using during the past month in Afghanistan aren't hi-res or high-bandwidth either, but that didn't make them useless.
Can this be the, uh, future?
No, not if it's existed for decades. It's what's referred to as a "mainframe". You know. An expandable number of processor boards running under an operating system that can treat them as any number of single-processor or multiprocessor machines, with the ability to reassign processes between CPUs.
The Unix world has had them for a long time, too. Modern examples include Sun's higher-end servers, which support hot-swappable and hot-pluggable processors and memory.
Doing it with x86 processors and standard x86 OSes like x86 Unixes and Windows is less common but I believe Compaq and maybe Unisys can sell you machines that can do it, too, with one or several instances of the OS running at once.
This hatdware approach is not quite the same as VMWare's server products, which do it via software and don't limit you to one OS per processor or block of processors. It in turn mimics other decades-old mainframe operating environments in approach.