NTFS has data checksums to detect and repair corruption caused by any component? (emphasis mine) No file system can repair all random data corruption, even ZFS. Among those that can always detect it and will generally fix it, however, is NTFS. (yawn)
Other things in NTFS and ZFS have in common:
Dynamic volumes. Efficient journaling. Useful, fast compression. Automatic periodic snapshots in 2k3 and Vista. Variable block sizes (oh, boy).
Open source? Not in the slightest.
But then, ZFS is so fucking open that it's license is incompatible with the Linux kernel and also, I expect, anything else with the GNU. So, as a Linux user, I don't count that as being too detrimental -- I can't use NTFS -or- ZFS.
The latter points are less about the functional output of Roland but rather an attempt to outline the emotional response to that output. My use of the word "whether" was intended to separate these issues, and indeed to focus on this emotional aspect.
Furthermore, I believe that this sort of synopsis may be beneficial to a discussion, at least when there are human beings doing the discussing.
The beef is that he is his own personal shill. Nearly every story he submits is a link to his own blog.
Whether they're interesting stories or not, and whether his stories are worse than having no Roland at all, it's the sort of blatant self-promotion that people on Slashdot are finely attuned toward hating. It is an affront to the sort of chaotic diversity that we've grown accustomed to having here, and folks don't like it.
(This comment is probably too late to be read by any meaningful number of people. Oh well.)
As for "free software is anti-business" and "companies can't make money off their work" - bullshit. Completely agreed. But I think you're missing a part of the more grand scheme here: Who gives a shit if there is money to be made?
I, for one, am completely fucking tired of people supporting the supposition that a thing is only successful if there is money being made from it, or that it is popular.
From my own perspective, I've been using free software on the desktop and on servers and wherever else I find it applicable for about the last 13 years.
13 years ago, it was advanced and worked well compared to other things of the time (Windows 3, DOS, OS/2, MacOS), but it wasn't everywhere. Linux-based devices were rare or unheard of. There was little money to be made from it.
13 years hence, it is still advanced and works well compared to other things of today (Windows Vista, XP, osX), and it's all over the place.
It is therefore successful. Not necessarily monetarily successful and popularly successful, though it sometimes (and increasingly) is. The important part of free software's success is that it remains usefully functional.
Free software doesn't care how much money it makes, or how many times it is copied and installed, or how often its license is completely exploited. Nor should the users of it. All that it needs to live and thrive is, really, that it remain free and that it still performs useful work on par with other industry offerings.
It doesn't need money to survive. It doesn't need popularity. The majority of the best and most-functional free software is (or was) written by certain individuals, driven by personal desire and need, who never stand to earn a dime from it. And it works just as well whether it is used by a hundred people or a million.
It's not just you. I wish the concept of caddy-loaded CD-ROMs would have caught on more than it did.
However, I've found SD to be at least a little bit resilient.
Anecdote: A few months ago, I had a 256MB Sandisk card and USB card reader sticking out of the front USB port on a computer, which was of course sitting on the floor. I kicked it rather hard (quite by accident), which scattered the combination across the room in several pieces.
So I cursed and swore and made some noise about the pain in my foot, and then went to gather up the various parts. It turns out that the reader works fine, though the guide rails are a bit smashed. And the SD card itself also works OK, despite its gruesome appearance with the housing mostly peeled off of the circuit board.
I don't trust either one of them for daily use at this point because they're no longer mechanically sound, but I was able to rescue the data on that card without any additional trauma. I have no doubt that the reader and the card would make a full recovery given some careful application of epoxy, but it was easier in this case to replace them.
The card may find a home yet, hardwired into a WRT54G with OpenWRT. The only thing preventing me from doing that is that I have no earthly idea what I would use an additional quarter-gig of storage for on my cable modem router.:)
Also, CF has pin-for-pin compatibility with normal 40-pin IDE/PATA devices, which is just plain cool. It's not unheard of to use a CF card as a solid-state disk in a router or other linux-embedded device. A very handy feature, indeed. I do wonder, though, if the SD-to-CF adapters are sufficiently compatible with IDE so as to accomplish the same trick.
But embedded Linux routers? I'm not very impressed. I've got Windows XP running 24/7 on a 2-gigabyte CF card.[1][4]
[1]: No, I'm not kidding. It works fine.[2] [2]: Yes, even the swapfile is on flash.[3] [3]: With wear-leveling being performed by card, I don't anticipate this being a problem within my lifetime. [4]: I'd switch the box over to Linux in an instant, if there were a method of programming the EMU10k1 DSP chip under Linux which were even 1/100th as convenient and flexible as the free kX drivers under Windows. But, sadly, there is not.
Such adapters exist, and aren't too hard to find. Here is an example.
Things like this will keep CF around for a bit longer, but I do suspect that its days are numbered. Flash is currently improving faster than CF-sized hard drives, so the little disks which made CompactFlash so desirable as a pro standard are no longer important.
And, there's something about the big, fat, durable, and mostly self-cleaning contacts on an SD card which makes the insertion process a whole lot less scary than the 40 pin (!!!) socket connector of CF.
Other than that, it's just a lot more compatible. My PDA, laptop, cell phone, car stereo, and consumer digital camera all have SD slots on them.
I'll miss CF when its gone, though, because the format's inherent ability to act, pin-for-pin, just like IDE hard drives makes for some useful (though probably not very interesting) hacks, which is something that none of the other flash formats are currently capable of. I've currently got two diskless computers here booting directly from CompactFlash cards which are plugged directly into the IDE bus, which has so far worked quite nicely. One is an old 386 laptop which now has zero moving parts (and which should last indefinitely), while the other is a K6-2 box that is doing some audio DSP work (which is now almost silent).
I suppose it might be about me, and not some general hypothetical case. Then again, it might not be.
It doesn't matter.
I wouldn't hire you to mop the floors, let alone work on computers. You've demonstrated a complete lack of respect for the most rudimentary project specifications and clear-cut business decisions. You insist that the company must change to better fit the available calender products, instead of the calender products changing to fit the company.
It has to work across multiple platforms because we have multiple platforms in use.
And since an on-site solution apparently does not exist (else you would have proposed something, as requested), I submit that you're just arguing for the sake of being difficult.
It is unfortunate, indeed, that the best tool for the job seems to be an off-site solution managed by someone else. Be it unfavorable and insipid, it is still the most productive and useful.
Firstly, while it is plain that there are already calendar synchronization applications for mobile devices, it is not plain that any of them actually fucking work across multiple platforms with a meaningful featureset. If you know differently, then please tell me about it, as from what I can see the market is full of festering shit.
Secondly, it is obvious that you're not in small business. The choice between paying through the nose for a T1 with an SLA in small town Ohio, or converting to pen-and-paper for the few hours-per-year that the cable modem drops out is obvious. As a business decision, the most profitable solution will (or at least should) always win. Cable modem it is.
Thirdly, I'm not "hosting it on comcast cable!" You've heard of Google, no? I'd guess that they've got better connectivity than Comcast.
Better get back under your rock, now. The sun will come up soon.
Suppose that I, for example, run a small service-oriented business, with technicians and service trucks and customer appointments, in addition to the normal gamut of meetings and other internals. Suppose that some of these technicians are located in different towns.
Suppose that, because of geographic disparity, it becomes a pain in the ass to manage a central paper-based schedule and keep everyone on the same page. And suppose I find that the solution to this problem is to implement some sort of network-aware calender. And, that I want to be able to access and modify this calender by a variety of means, from standalone PalmOS devices to Windows boxen to WinCE phones, because the different techs, salespeople, and managers all have their own levels of technical ability and devices of choice.
And now, just suppose that something like Google Calender fits this bill and is put in service. Everyone knows where everyone else is, what they're doing later today (or next week). Scheduling a job can happen easily, and conflicts can be seen and avoided immediately. Life is good, and the paper schedule is forgotten (good riddance).
With me so far?
Good.
Now, suppose that the Intar-web is down, and Google Calender is unreachable.
Trucks stop rolling. Customers get angry about missed appointments. Jobs don't get done. And, the kicker: Nobody, except perhaps the stubborn old geek with an offline Palm Pilot, has any idea what anyone (including themselves!) is supposed to be doing. The company basically takes a vacation until connectivity is restored, which (in small business) means waiting as long as it takes for Time Warner or SBC to correct the problem.
Having offline web application support, if implemented well, can fix this problem. Even if new jobs can't be scheduled electronically, at least work on existing stuff can continue, as all that it takes is one person with Firefox on a desktop machine to pass out orders.
The worst-case, then, goes from having no data at all and a complete cessation of work, to at least having old data. A notepad and cell phones can then fill in the blanks for new jobs (just like it used to), which can be entered into the calender system once the Internet connection comes back.
Better idea: Just rescue a Pentium-class machine from the curb and be done with it. I know it's so obvious that it hurts, but *come on*, man.
And then, if you still need/want extra points, remove as many of the critical moving parts from the box as you can to enhance reliability. Think undervolting, big heatsinks, and solid-state storage.
But $180 is too much to spend just for geek cred alone.
If that's your whole goal, then look not toward needless complication. Far better (and cheaper) results would come cheaper from a $50 WRT54G-ish router and running the audio over the network or perhaps out through an attached USB sound card. 216MHz of 32-bit ARM should decode OGG mightily, I'd guess, based on the fair job my old Riovolt does with OGG with its much slower 75MHz chip.
I've noticed the trend toward "No problem," myself. But even that is at least civil. Lately, just in the past year or so, I've been hearing "Yep" with alarming frequency after I tell someone thanks.
Example:
Waitress: Would you like another [beer/coffee/Coke]? Me: Please. *brief pause* *beverage lands on table* Me: Thank you. Waitress: Yep.
Needless to say, this is not good for the income levels of people who rely on tips to stay alive.
At work, in the past six months, I've replaced/discarded the following routers, due to each one of them showing signs that it was no longer properly functional or reliable:
With my Verizon phone, pressing 6 during any of the preambles will skip to the next step.
Since discovering this, my voice mail listening sessions have been shortened to:
1. Press VM hotkey. 2. Wait for for the welcoming speech to begin. 3. Interrupt welcome speech by immediately entering passcode. 4. Interrupt speech detailing my mailbox contents by mashing 6. 5. Interrupt pedantic lecture detailing the header information for the upcoming voicemail by pressing 6. 6. Listen to voice mail message. 7. Press 7 to delete it. 8. Close flip.
Takes only about 30 seconds, now. It used to be much worse.
When I was a kid, we had a big combination range/oven with an integrated, overhead microwave, branded Kenmore. It had a timer knob, and a power knob, a start button, and a door open button. The timer wasn't logarithmic, but did have two zones where the friction/gearing/whatever was different: IIRC, the first half of the knob's range of motion was good up to about 15 minutes, and the last half was harder to turn through, and would let it run for up to an hour. There was a mechanical bell, activated as the timer expired, which sounded a single, very clear "Ding!" to announce that it was done.
It still exists, and works fine. I know where it lives. And: I'd be happy to broker the purchase for you, if you'd be interested in procuring such an antiquity.
Monitors:
I do not know what you're after, exactly. It sounds like you want either easily-selectable presets, or just a way to boost brightness temporarily.
My ViewSonic P95f+ CRT has a magic button on the front panel (easily identified by being the left-most button) which vastly increases brightness. Three modes: "NORMAL:TEXT/SPREADSHEET", "ULTRABRITE 3x:GRAPHICS/GAME", and "ULTRABRITE 4x:VIDEO/MOVIE/DVD". I've used it about twice in the three years I've had the monitor; it's not very appealing to me...
Gee, thanks. I'd never thought of it that way./me awaits onslaught of cheap, nasty Lexmark printers capable of printing with millions-of-colors by way of spacial dithering, in contrast to the competition's flagship slovenly 6-color models.
Even if we assume that your figures are correct: Your argument only makes sense if one assumes that the 16,000 colors being displayed by the computer monitor are perfectly aligned with the 16,000 colors capable of being seen by the eye.
This is, of course, most certainly not the case, as they're both nonlinear systems with completely different sets of limiting factors, and they overlap poorly.
Besides: I'm sure that I'm not the only one on Earth who is able to see visible banding in 15- and 16-bit color modes (which are of course 32 and 64k colors, respectively). It's not something I notice all the time, but it does show up occasionally with everything from photos, to movies, to stupid Powerpoint slides. In those instances, switching to the display to 24-bit ("millions of colors") mode has usually cured the problem, for me.
In conclusion: 16-bit banding is annoying and distracting and therefore inferior and plainly different to 24-bit color*.
[*]: I'd proceed with arguing for the need to move display devices beyond 24-bit color, which is also not without its own pitfalls, but I'm too tired to bother just now.
WiMAX, despite all of the current buzz, is completely unsuitable for the sort of impromptu wireless network access that municipalities are attempting to provide.
The frequencies are all wrong, the standards are too fragmented, and mass-market hardware has been "right around the corner" long enough to make a timely release of Duke Nuken Forever look like a realistic forecast.
Quoth Wikipedia:
The original WiMAX standard (IEEE 802.16) specified WiMAX for the 10 to 66 GHz range. 802.16a, updated in 2004 to 802.16-2004, added specifications for the 2 to 11 GHz range. 802.16-2004 was updated to 802.16e in 2005 and uses scalable orthogonal frequency-division multiple access (SOFDMA) as opposed to the OFDM version with 256 sub-carriers (of which 200 are used) in 802.16d. So the very best possible case is that we (in the US) get to use WiMax in the 2.4GHz ISM spectrum already crowded by cordless phones, baby monitors, the neighbors' BitTorrent-leeching laptops, the local McDonalds, and the peeping tom's X10 pottycam. Just like we do right now, with WiFi. We even get to keep the same FCC-mandated power limits (yay!), which neither money nor municipality status will permit exception to; only certain license-classes of HAM operators are allowed to exceed the standard ISM radiated power limits at 2.4GHz. (Advanced modulation techniques will help a bit, but for that we also have 802.11n, also due Real Soon Now.)
Higher frequencies don't penetrate buildings or trees for shit, so count out the 5.8GHz band, or anything else license-free within the ranges specified above (as plainly evidenced by the prolific lack of 802.11a in the field). The tens-of-GHz bands all suffer obstruction limitations similar to that of visible light: If you can't see the access point, you can't connect to it at any meaningful range.
That said, the 700MHz (quasi-former UHF NTSC TV) band is very promising, but the FCC is handling it in such a fashion so as to nearly certify its uselessness. It penetrates buildings and trees quite well, and higher power is permitted. But: Having personally interviewed a number of different vendors of actually-extant 700MHz gear, none of them seem to be taking WiMax (or any other interoperable standard) seriously in that band. They'd all rather stay proprietary.
Furthermore, as someone who consults with a 700MHz license holder who someday hopes to cash in on the band, the 12MHz of total bandwidth allocated by the FCC to each (bloody expensive) license is grossly inadequate to deliver fast network access to the teaming masses without a very dense population of (bloody expensive) access points.
IMHO, Metricom had a better idea with their inexpensive 900MHz network of several years ago, and they still failed miserably.
So: I'm sorry to report that WiMax, as things stand, is really only potentially useful for tower-to-tower communications. Think backhauls for cellular or localized 802.11a/b/g/n WiFi networks, or maybe a link between campuses, or between corporate buildings. It'd be nice if it were useful for end-users, but it's really not.
Usually nobody worth his salt uses electrolytes for timing
Naah. They're used everywhere that analog filters are -- look at nearly any >$5 sound card, near the analog IO. There's dozens of the little buggers.
And no, I haven't checked, but I'd bet there's a good assortment of ratty electrolytics near the F or RJ11 jack of a cable or DSL modem, too, all subject to whatever transients might transpire on the cable TV / telephone networks.
And, that line is likely to have either no secondary surge suppression other than what is provided at the demarc, or secondary surge suppression which is very often worse than none at all: An MOV to the electrical outlet's ground, with MOVs from ground to line and neutral. This scares the hell out of me, because if that ground reference is floating, or high-impedance, or otherwise imperfect (ie: typical), the circuit will gladly conduct the surge from wherever it is, to wherever else you don't want it to go.
IE: With a poorly-grounded outlet, a nearby lightning strike on an electric wire will easily jump over to the cable modem's coax. This transition happens right next to the computer, courtesy of the "surge suppressor," so it has a nice low-impedance path straight into the modem.
So my money's still on the front end being partially hosed.
Electronics (specifically, the electrolytic capacitors which inundate consumer electronics) don't last forever. Eventually, they dry up a bit. This causes their characteristics to change, and never in a good way.
Eventually, they'll fail completely. Depending on the nature of the failure, and the design of the circuit, they may or may not destroy other components at the same time.
Could take months, years, or many decades. The changes to capacitance and ESR will be most likely to affect analog circuitry first, and probably in somewhat subtle ways (like, say, the RF front end of a cable modem degrading slowly over time, showing poor signal levels and/or connectivity issues but while generally still working OK).
[I often butt heads with the repair techs where I work. They insist that there is nothing more damaging to electronics than to leave them powered on all the time and walk around turning things off, while I proclaim that there's no point in even having network-accessible electronics if they are always switched off every time you want to use them. *sigh*]
Slashdot Mirror
Other things in NTFS and ZFS have in common:
Dynamic volumes. Efficient journaling. Useful, fast compression. Automatic periodic snapshots in 2k3 and Vista. Variable block sizes (oh, boy).
Open source? Not in the slightest.
But then, ZFS is so fucking open that it's license is incompatible with the Linux kernel and also, I expect, anything else with the GNU. So, as a Linux user, I don't count that as being too detrimental -- I can't use NTFS -or- ZFS.
I see your point. Please allow me to explain.
The latter points are less about the functional output of Roland but rather an attempt to outline the emotional response to that output. My use of the word "whether" was intended to separate these issues, and indeed to focus on this emotional aspect.
Furthermore, I believe that this sort of synopsis may be beneficial to a discussion, at least when there are human beings doing the discussing.
The beef is that he is his own personal shill. Nearly every story he submits is a link to his own blog.
Whether they're interesting stories or not, and whether his stories are worse than having no Roland at all, it's the sort of blatant self-promotion that people on Slashdot are finely attuned toward hating. It is an affront to the sort of chaotic diversity that we've grown accustomed to having here, and folks don't like it.
I, for one, am completely fucking tired of people supporting the supposition that a thing is only successful if there is money being made from it, or that it is popular.
From my own perspective, I've been using free software on the desktop and on servers and wherever else I find it applicable for about the last 13 years.
13 years ago, it was advanced and worked well compared to other things of the time (Windows 3, DOS, OS/2, MacOS), but it wasn't everywhere. Linux-based devices were rare or unheard of. There was little money to be made from it.
13 years hence, it is still advanced and works well compared to other things of today (Windows Vista, XP, osX), and it's all over the place.
It is therefore successful. Not necessarily monetarily successful and popularly successful, though it sometimes (and increasingly) is. The important part of free software's success is that it remains usefully functional.
Free software doesn't care how much money it makes, or how many times it is copied and installed, or how often its license is completely exploited. Nor should the users of it. All that it needs to live and thrive is, really, that it remain free and that it still performs useful work on par with other industry offerings.
It doesn't need money to survive. It doesn't need popularity. The majority of the best and most-functional free software is (or was) written by certain individuals, driven by personal desire and need, who never stand to earn a dime from it. And it works just as well whether it is used by a hundred people or a million.
I don't see that changing.
Rather than suffer through more than 100 pages (WTF?) of advertisements and bad testing, I'd like to refer all of you to this simple page:
The Dan's Data CPU Cooler Snap Judgement Guide
It's about five years old, but the thermodynamic problem of removing waste heat from an object is about the same as it has ever been.
It's not just you. I wish the concept of caddy-loaded CD-ROMs would have caught on more than it did.
:)
However, I've found SD to be at least a little bit resilient.
Anecdote: A few months ago, I had a 256MB Sandisk card and USB card reader sticking out of the front USB port on a computer, which was of course sitting on the floor. I kicked it rather hard (quite by accident), which scattered the combination across the room in several pieces.
So I cursed and swore and made some noise about the pain in my foot, and then went to gather up the various parts. It turns out that the reader works fine, though the guide rails are a bit smashed. And the SD card itself also works OK, despite its gruesome appearance with the housing mostly peeled off of the circuit board.
I don't trust either one of them for daily use at this point because they're no longer mechanically sound, but I was able to rescue the data on that card without any additional trauma. I have no doubt that the reader and the card would make a full recovery given some careful application of epoxy, but it was easier in this case to replace them.
The card may find a home yet, hardwired into a WRT54G with OpenWRT. The only thing preventing me from doing that is that I have no earthly idea what I would use an additional quarter-gig of storage for on my cable modem router.
But embedded Linux routers? I'm not very impressed. I've got Windows XP running 24/7 on a 2-gigabyte CF card.[1][4]
[1]: No, I'm not kidding. It works fine.[2]
[2]: Yes, even the swapfile is on flash.[3]
[3]: With wear-leveling being performed by card, I don't anticipate this being a problem within my lifetime.
[4]: I'd switch the box over to Linux in an instant, if there were a method of programming the EMU10k1 DSP chip under Linux which were even 1/100th as convenient and flexible as the free kX drivers under Windows. But, sadly, there is not.
Such adapters exist, and aren't too hard to find. Here is an example.
Things like this will keep CF around for a bit longer, but I do suspect that its days are numbered. Flash is currently improving faster than CF-sized hard drives, so the little disks which made CompactFlash so desirable as a pro standard are no longer important.
And, there's something about the big, fat, durable, and mostly self-cleaning contacts on an SD card which makes the insertion process a whole lot less scary than the 40 pin (!!!) socket connector of CF.
Other than that, it's just a lot more compatible. My PDA, laptop, cell phone, car stereo, and consumer digital camera all have SD slots on them.
I'll miss CF when its gone, though, because the format's inherent ability to act, pin-for-pin, just like IDE hard drives makes for some useful (though probably not very interesting) hacks, which is something that none of the other flash formats are currently capable of. I've currently got two diskless computers here booting directly from CompactFlash cards which are plugged directly into the IDE bus, which has so far worked quite nicely. One is an old 386 laptop which now has zero moving parts (and which should last indefinitely), while the other is a K6-2 box that is doing some audio DSP work (which is now almost silent).
I suppose it might be about me, and not some general hypothetical case. Then again, it might not be.
It doesn't matter.
I wouldn't hire you to mop the floors, let alone work on computers. You've demonstrated a complete lack of respect for the most rudimentary project specifications and clear-cut business decisions. You insist that the company must change to better fit the available calender products, instead of the calender products changing to fit the company.
You are an imbecile.
Hope this helps.
It has to work across multiple platforms because we have multiple platforms in use.
And since an on-site solution apparently does not exist (else you would have proposed something, as requested), I submit that you're just arguing for the sake of being difficult.
It is unfortunate, indeed, that the best tool for the job seems to be an off-site solution managed by someone else. Be it unfavorable and insipid, it is still the most productive and useful.
Firstly, while it is plain that there are already calendar synchronization applications for mobile devices, it is not plain that any of them actually fucking work across multiple platforms with a meaningful featureset. If you know differently, then please tell me about it, as from what I can see the market is full of festering shit.
Secondly, it is obvious that you're not in small business. The choice between paying through the nose for a T1 with an SLA in small town Ohio, or converting to pen-and-paper for the few hours-per-year that the cable modem drops out is obvious. As a business decision, the most profitable solution will (or at least should) always win. Cable modem it is.
Thirdly, I'm not "hosting it on comcast cable!" You've heard of Google, no? I'd guess that they've got better connectivity than Comcast.
Better get back under your rock, now. The sun will come up soon.
Suppose that I, for example, run a small service-oriented business, with technicians and service trucks and customer appointments, in addition to the normal gamut of meetings and other internals. Suppose that some of these technicians are located in different towns.
Suppose that, because of geographic disparity, it becomes a pain in the ass to manage a central paper-based schedule and keep everyone on the same page. And suppose I find that the solution to this problem is to implement some sort of network-aware calender. And, that I want to be able to access and modify this calender by a variety of means, from standalone PalmOS devices to Windows boxen to WinCE phones, because the different techs, salespeople, and managers all have their own levels of technical ability and devices of choice.
And now, just suppose that something like Google Calender fits this bill and is put in service. Everyone knows where everyone else is, what they're doing later today (or next week). Scheduling a job can happen easily, and conflicts can be seen and avoided immediately. Life is good, and the paper schedule is forgotten (good riddance).
With me so far?
Good.
Now, suppose that the Intar-web is down, and Google Calender is unreachable.
Trucks stop rolling. Customers get angry about missed appointments. Jobs don't get done. And, the kicker: Nobody, except perhaps the stubborn old geek with an offline Palm Pilot, has any idea what anyone (including themselves!) is supposed to be doing. The company basically takes a vacation until connectivity is restored, which (in small business) means waiting as long as it takes for Time Warner or SBC to correct the problem.
Having offline web application support, if implemented well, can fix this problem. Even if new jobs can't be scheduled electronically, at least work on existing stuff can continue, as all that it takes is one person with Firefox on a desktop machine to pass out orders.
The worst-case, then, goes from having no data at all and a complete cessation of work, to at least having old data. A notepad and cell phones can then fill in the blanks for new jobs (just like it used to), which can be entered into the calender system once the Internet connection comes back.
Which is quite likely good enough.
I, for one, welcome our fascist overlords!
Better idea: Just rescue a Pentium-class machine from the curb and be done with it. I know it's so obvious that it hurts, but *come on*, man.
And then, if you still need/want extra points, remove as many of the critical moving parts from the box as you can to enhance reliability. Think undervolting, big heatsinks, and solid-state storage.
But $180 is too much to spend just for geek cred alone.
If that's your whole goal, then look not toward needless complication. Far better (and cheaper) results would come cheaper from a $50 WRT54G-ish router and running the audio over the network or perhaps out through an attached USB sound card. 216MHz of 32-bit ARM should decode OGG mightily, I'd guess, based on the fair job my old Riovolt does with OGG with its much slower 75MHz chip.
I've noticed the trend toward "No problem," myself. But even that is at least civil. Lately, just in the past year or so, I've been hearing "Yep" with alarming frequency after I tell someone thanks.
Example:
Waitress: Would you like another [beer/coffee/Coke]?
Me: Please.
*brief pause*
*beverage lands on table*
Me: Thank you.
Waitress: Yep.
Needless to say, this is not good for the income levels of people who rely on tips to stay alive.
At work, in the past six months, I've replaced/discarded the following routers, due to each one of them showing signs that it was no longer properly functional or reliable:
1. "Network Anywhere" wired router
1. Linksys BEFSR41
1. Linksys BEFW11S4
1. ZyXEL P-660H
It should be noted that, company-wide, we use a total of seven consumer-grade routers and that, as noted above, four of them have recently failed.
For whatever it's worth.
I cannot speak about other providers, but:
With my Verizon phone, pressing 6 during any of the preambles will skip to the next step.
Since discovering this, my voice mail listening sessions have been shortened to:
1. Press VM hotkey.
2. Wait for for the welcoming speech to begin.
3. Interrupt welcome speech by immediately entering passcode.
4. Interrupt speech detailing my mailbox contents by mashing 6.
5. Interrupt pedantic lecture detailing the header information for the upcoming voicemail by pressing 6.
6. Listen to voice mail message.
7. Press 7 to delete it.
8. Close flip.
Takes only about 30 seconds, now. It used to be much worse.
Microwaves:
When I was a kid, we had a big combination range/oven with an integrated, overhead microwave, branded Kenmore. It had a timer knob, and a power knob, a start button, and a door open button. The timer wasn't logarithmic, but did have two zones where the friction/gearing/whatever was different: IIRC, the first half of the knob's range of motion was good up to about 15 minutes, and the last half was harder to turn through, and would let it run for up to an hour. There was a mechanical bell, activated as the timer expired, which sounded a single, very clear "Ding!" to announce that it was done.
It still exists, and works fine. I know where it lives. And: I'd be happy to broker the purchase for you, if you'd be interested in procuring such an antiquity.
Monitors:
I do not know what you're after, exactly. It sounds like you want either easily-selectable presets, or just a way to boost brightness temporarily.
My ViewSonic P95f+ CRT has a magic button on the front panel (easily identified by being the left-most button) which vastly increases brightness. Three modes: "NORMAL:TEXT/SPREADSHEET", "ULTRABRITE 3x:GRAPHICS/GAME", and "ULTRABRITE 4x:VIDEO/MOVIE/DVD". I've used it about twice in the three years I've had the monitor; it's not very appealing to me...
Gee, thanks. I'd never thought of it that way. /me awaits onslaught of cheap, nasty Lexmark printers capable of printing with millions-of-colors by way of spacial dithering, in contrast to the competition's flagship slovenly 6-color models.
Even if we assume that your figures are correct: Your argument only makes sense if one assumes that the 16,000 colors being displayed by the computer monitor are perfectly aligned with the 16,000 colors capable of being seen by the eye.
This is, of course, most certainly not the case, as they're both nonlinear systems with completely different sets of limiting factors, and they overlap poorly.
Besides: I'm sure that I'm not the only one on Earth who is able to see visible banding in 15- and 16-bit color modes (which are of course 32 and 64k colors, respectively). It's not something I notice all the time, but it does show up occasionally with everything from photos, to movies, to stupid Powerpoint slides. In those instances, switching to the display to 24-bit ("millions of colors") mode has usually cured the problem, for me.
In conclusion: 16-bit banding is annoying and distracting and therefore inferior and plainly different to 24-bit color*.
[*]: I'd proceed with arguing for the need to move display devices beyond 24-bit color, which is also not without its own pitfalls, but I'm too tired to bother just now.
The frequencies are all wrong, the standards are too fragmented, and mass-market hardware has been "right around the corner" long enough to make a timely release of Duke Nuken Forever look like a realistic forecast.
Quoth Wikipedia: The original WiMAX standard (IEEE 802.16) specified WiMAX for the 10 to 66 GHz range. 802.16a, updated in 2004 to 802.16-2004, added specifications for the 2 to 11 GHz range. 802.16-2004 was updated to 802.16e in 2005 and uses scalable orthogonal frequency-division multiple access (SOFDMA) as opposed to the OFDM version with 256 sub-carriers (of which 200 are used) in 802.16d. So the very best possible case is that we (in the US) get to use WiMax in the 2.4GHz ISM spectrum already crowded by cordless phones, baby monitors, the neighbors' BitTorrent-leeching laptops, the local McDonalds, and the peeping tom's X10 pottycam. Just like we do right now, with WiFi. We even get to keep the same FCC-mandated power limits (yay!), which neither money nor municipality status will permit exception to; only certain license-classes of HAM operators are allowed to exceed the standard ISM radiated power limits at 2.4GHz. (Advanced modulation techniques will help a bit, but for that we also have 802.11n, also due Real Soon Now.)
Higher frequencies don't penetrate buildings or trees for shit, so count out the 5.8GHz band, or anything else license-free within the ranges specified above (as plainly evidenced by the prolific lack of 802.11a in the field). The tens-of-GHz bands all suffer obstruction limitations similar to that of visible light: If you can't see the access point, you can't connect to it at any meaningful range.
That said, the 700MHz (quasi-former UHF NTSC TV) band is very promising, but the FCC is handling it in such a fashion so as to nearly certify its uselessness. It penetrates buildings and trees quite well, and higher power is permitted. But: Having personally interviewed a number of different vendors of actually-extant 700MHz gear, none of them seem to be taking WiMax (or any other interoperable standard) seriously in that band. They'd all rather stay proprietary.
Furthermore, as someone who consults with a 700MHz license holder who someday hopes to cash in on the band, the 12MHz of total bandwidth allocated by the FCC to each (bloody expensive) license is grossly inadequate to deliver fast network access to the teaming masses without a very dense population of (bloody expensive) access points.
IMHO, Metricom had a better idea with their inexpensive 900MHz network of several years ago, and they still failed miserably.
So: I'm sorry to report that WiMax, as things stand, is really only potentially useful for tower-to-tower communications. Think backhauls for cellular or localized 802.11a/b/g/n WiFi networks, or maybe a link between campuses, or between corporate buildings. It'd be nice if it were useful for end-users, but it's really not.
Usually nobody worth his salt uses electrolytes for timing
:)
Naah. They're used everywhere that analog filters are -- look at nearly any >$5 sound card, near the analog IO. There's dozens of the little buggers.
And no, I haven't checked, but I'd bet there's a good assortment of ratty electrolytics near the F or RJ11 jack of a cable or DSL modem, too, all subject to whatever transients might transpire on the cable TV / telephone networks.
And, that line is likely to have either no secondary surge suppression other than what is provided at the demarc, or secondary surge suppression which is very often worse than none at all: An MOV to the electrical outlet's ground, with MOVs from ground to line and neutral. This scares the hell out of me, because if that ground reference is floating, or high-impedance, or otherwise imperfect (ie: typical), the circuit will gladly conduct the surge from wherever it is, to wherever else you don't want it to go.
IE: With a poorly-grounded outlet, a nearby lightning strike on an electric wire will easily jump over to the cable modem's coax. This transition happens right next to the computer, courtesy of the "surge suppressor," so it has a nice low-impedance path straight into the modem.
So my money's still on the front end being partially hosed.
Too bad we'll never know.
I, for one, welcome our chair-throwing overlords!
Electronics (specifically, the electrolytic capacitors which inundate consumer electronics) don't last forever. Eventually, they dry up a bit. This causes their characteristics to change, and never in a good way.
Eventually, they'll fail completely. Depending on the nature of the failure, and the design of the circuit, they may or may not destroy other components at the same time.
Could take months, years, or many decades. The changes to capacitance and ESR will be most likely to affect analog circuitry first, and probably in somewhat subtle ways (like, say, the RF front end of a cable modem degrading slowly over time, showing poor signal levels and/or connectivity issues but while generally still working OK).
[I often butt heads with the repair techs where I work. They insist that there is nothing more damaging to electronics than to leave them powered on all the time and walk around turning things off, while I proclaim that there's no point in even having network-accessible electronics if they are always switched off every time you want to use them. *sigh*]
Your entire argument is worthless. You should have thought it out first.
I agree.
Thank you for pointing that out.