Note the site says explicitly that they're in "early stages of development", and one thing they hope to achieve is real-time decode. Note they don't even mention real-time encode, which is probably an order of magnitude harder.
It does sound interesting, however as people here have noted a number of techniques it apparently uses (such as arithmetic coding, probably wavelets in general, etc) are known to be patented. H.264 Baseline Profile doesn't use arithmetic coding (CABAC) because Baseline is designed to avoid most/all of the patent-licensing issues. H.264 Main Profile uses CABAC, etc (and gets better compression because of it).
Back in the day (Amiga 3000 introduction circa 1990) we could boot an Amiga off a 40MB disk from power-on (including loading a replacement for the ROM image) to fully up (including mounting NFS mounts) in ~15 seconds; 7 seconds from a warm boot. Effectively we chunked the load of much of the OS (including GFX, Workbench (desktop), etc) as part of the softload of the ROM image into RAM. On top of that, on warm boot we checksummed the ROM image in RAM (as well as various major OS library structures), and if they were good we didn't reload them. (Remember, the Amiga ran without memory protection.) Some of the boot script that followed would cause overlap (anything tha spawned servers/processes), other parts were sequential but generally were very fast and often cached (like "ASSIGN XYZ: dh0:xyz" commands and the like).
At the time, Win3.1 took Quite A While to boot. Getting past the BIOS alone usually took longer than the Amiga took to finish booting.
Trying to do all the reordering in the OS (as suggested in several posts here) seems like a good idea, but ignores some issues:
Disks aren't a uniform array of blocks, and even if you have disk geometry it's almost certainly at least simplified, and probably a total lie. (you can query a SCSI drive for the next "slowdown", but what that means is ill-defined, and not that useful anyways.)
Because of (1), you don't know when blocks are on the same track or not.
it may be more efficient head-movement-wise not to switch directions.
When reordering requests, head position matters, but so does rotational angle, and the OS scheduler has no idea about that.
Hardware raid. Now you really don't know what the geometry is or what the IO scheduling parameters are. Not to mention letting the raid controller make use of multiple drives efficiently for small requests.
This PDF shows that tagged queuing in SCSI drives is a win even with host IO ordering, when the IO stream is random or fairly real-world (lots of independant streams of reads and writes). There's minimal or no win for sequential IO.
Tagged queuing in SATA/etc drives is a step in the right direction, though last I checked it wasn't equal to TQ in SCSI. Native Command Queuing in SATA will probably give similar performance to TQ in SCSI.
ATA is definitely not SCSI-over-IDE. ATAPI is SCSI-over-IDE however.
I wrote the IDE/ATA drivers for the Amiga. The Amiga SCSI drivers accepted "SCSIDirect" commands from applications. Internally, all IO commands were converted to SCSIDirect commands for execution. To implement ATA, I added a SCSIDirect->ATA translator (which wasn't that hard - about 3 weeks from start to working, booting system - and I implemented just about all SCSI commands even semi-reasonable (all of CCS I think, plus quite a bit).
Doing it this way made implementing support for ATAPI CDROMs (something I did as a contract after Commodore folded) Very Easy.:-)
I'm fairly certain (having been on the Amiga OS team at the time, and having worked with Bill a fair bit) that he was not stiffed by Commodore on anything. I don't know what sort of deal there was for including ARexx in the OS - it might not have included money.
Bill did a fair bit of work for Commodore on contract, and also did a lot of testing and tool development for us (evil memory-allocation failure tools, for example). We tried to hire him on multiple occasions, but he preferred to stay in Boston.
Bill seems to be involved in Linux kernel/etc stuff nowadays from a quick google search.
Arexx.... yum (I was on the Commodore Amiga OS team at the time).
BTW, Gnus in Emacs also _totally_ integrates mail and news. Plus like all things emacs, it's totally configurable/modifiable. Though elisp is much more of a pain to program in the ARexx.
I know the author (college), and I can tell you he's on a number of mailing lists, including the extended-college-ilk list. Not to mention that he's a TV star...;-) You've probably watched him (and rooted for him) on TV if you're a slashdotter/techie.
Now the negatives:
* Commodore: Entered an industry well penetrated by apple, IBM, Tandy (back then) and company and tried to play along, didn't make it...
Commodore: Commodore was one of the early ones; they were out there with Radio Shack and Apple, long before IBM tried personal computers. In the mid-80's they sold over 14 million C-64's; at the time that was a stupendous number. Commodore's failing was of management and goals, mostly due to upper management who had no idea what a computer was (Mehdi Ali; a wall-street banker type) and insisted on micro-managing the company - tough when you don't understand the products.
The final blow was when in fall of '93 Mehdi decided to build a few 10's of thousands of the new (AA/AGA) machines (A1200, etc), and 300,000+ of the old chipset-based machines (A600). Needless to say, the old machines didn't move off the shelves very fast at Xmas, and that was the final nail.
There were other instances like that too. Mostly it was caused by not following up on successful products (C64, A500, to some extent A3000) and trying to milk them for too long. The A1200 was the right machine; it was just too late by a year or two. Engineering had it's issues too, in particular biting off more than we could chew on the total redesign of the chipset which was never quite finished ("AAA"), and not giving enough attention to the potential high-volume products, though in general engineering was pretty focused on them.
It's tough when the CEO won't let marketing talk to engineering directly, and insists all contact go through him and his cronies...
Disclaimer: I'm an ex-Commodore engineer from these times, and after bankruptcy was declared, we burnt Mehdi Ali in effigy in my backyard (literally).
The retailer bought the units from Replay with the 3-year-activation for $XXX, and was reselling them for ~$500.
When Replay decided to change their model, the retailer took the option of selling the units without the activation for $150, and getting a check back from Replay for circa $200 (on their old stock). When they made that deal, it was incumbent on them to sell them without activation in order to get the check from Replay. They (CC) sold some of them (not all) without telling the consumer that the activation had been removed, but they then went and told Replay (or had already told Replay) that those units had been sold without activation.
It doesn't matter what's sealed in the box if the retailer (correctly) tells you that the deal is different. The retailer changed the deal, but didn't tell (all) the consumers. Replay has no direct knowledge of which consumers the retailer sold to under the old model or the new model, it only knows that CC told Replay that all those units (at $150) were sold without activation.
Note again that some CC's correctly marked them, and others marked them sometime during the day. Also note that most Radio Shacks correctly marked them _and_ the boxes had stickers that said activation required.
Circuit City screwed up, bigtime. One way or another, they're liable (though perhaps only to refund purchasers their money, perhaps to Replay (i.e. circuit city doesn't get the $250ish from Replay, and Replay provides the service that CC sold them)).
CC will probably claim it can't determine which people were told and which weren't, and so will probably only offer to refund people's $150ish in exchange for the units back.
The seller is not a 3rd party to the sale, they're a 1st party. They are not merely an agent for Replay; they buy inventory from Replay, then sell that to consumers. Replay can't tell them to mark the stock down and remove the activation, and Replay can't go in and demand they change the stickers. Replay can make a deal with the retailer, though, for the retailer to remove the activation and get paid by Replay when they sell it or when the user activates it, etc. In that case, it's incumbent on the retailer to live up to that bargain, and so inform the buyers of what they're buying.
Remember, when you buy something in a store you're buying it from the retailer. The manufacturer has a requirement to honor the deal they sold it to the retailer under - but in this case the retailer and the manufacturer agreed to a new deal that involved the retailer remarking the unsold stock, and the retailer screwed it up.
The phone reps at Replay couldn't know that CC has screwed up, nor that CC had jumped the gun and dropped the price early.
You stated that Replay "did NOT tell them to take off the markings". There is considerable evidence that they did, but that CC didn't do a good job of getting the website people and all the store managers involved in getting word out and remarking the shelves/stock.
Fundamentally, the purchase and promise of service was by the retailer. They separately had purchased the units (originally with a promise of service) from Replay.
Replay had told them it was changing the pricing model, and so that the retailers didn't end up holding the bag, they told the retailers they could sell the units without service and Replay would refund to them most or all of the service contract price. Part of that deal was that they were to be sold without the service included (and mark the units clearly as such), and that the retailers would report back which units were sold without service. (I know none of this directly, but I believe the details are very close to what I've described.)
CC sold the units, and in many stores (and on the web) didn't tell the buyers that these were now without service (which was why there was such a discount, since CC was to get a check back from Replay for each unit sold). CC is liable in some manner: either to the customer to provide service (they shouldn't have told Replay they sold it without service when they sold it with service), to the customer for a full refund (to correct for the mistake), or they're liable to Replay (Replay ends up providing service anyways, but doesn't pay CC).
Consumers who bought units at Radio Shack generally are out of luck (unless an employee told them it came with service - a few did, but many of them apparently knew to tell people service was extra, and the boxes were apparently mostly marked "service activation required" etc). Those people were really gambling that Radio Shack and Replay had simply made a mistake and they'd slip through unnoticed.
IMHO, IANAL, etc.
Or to put it another way, for all that sit salivating for the next "oops" rebate/etc snafu:
Replay has also been adding features, such as one to mostly replace (and in some way surpass) commercial-skip - Replay 5000's/5500's can now skip to the next or previous show segment (which may conveniently skip over all the commercials without having to hit the 30-sec skip key 6 times, then back up a few times).
It's also handy for skipping to the next segment of a show when they're doing a report/etc you don't care about.
As for your legal comments - it's not clear here that those comments apply to Replay or to the retailer. Or, if you prefer, the retailers are in violation for telling replay that the consumers bought the product without the 3-year service, when the retailer failed to remove the 3-year-service stickers. (Note that almost all Radio Shack units did NOT have 3-year-service stickers, though some RS employees told people it was included, and that some CC's (especially by later in the day) had removed/X-ed-out the stickers and had signs up).
First of all, the beef would be with the seller, not directly with Replay. And they can simply refund your money and take it back; that's generally considered complete restitution in a case such as this. Even under your math, Replay doesn't owe anyone $616 or $449; at most (if your interpretation were correct) was 3 years of service. The cost of that service (if bought) is $299; of course it costs them less, but that service fee subsidizes the below-cost-of-goods sale prices that both TiVo and Replay are now selling at.
As others may have said, Replay told retailers to remove the 3-year stickers, etc before dropping the price on all current inventory. CC and Radio Shack didn't bother to do this, and didn't update their websites either. Replay had set the price change for around Dec 20th; CC and Radio Shack (to get extra Xmas sales I assume) jumped the gun and did it about a week early. That's not Replay's fault. And the reps called to ask about the price answered correctly so far as they knew - the retailers were supposed to wait to change the price and the terms.
All those FatWallet-ers who bought the units knew it was an unintentional mistake (and probably knew it was the retailer's mistake). They figured they were getting something for nothing (effectively).
Basically, Replay is getting screwed over here by CC and the Shack (who also screwed themselves). Many of those units (bought by FW-ers) will be returned, and have to be sold eventually as refurbs or open-box units. (Earlier this year, Replay sold refurbs with lifetime service for ~$300 (service was $250 then)).
CC and RS are telling Replay "cancel the activation on these serial numbers, we sold them without service, then send us a check" since CC/RS paid for that inventory when it did include service, but they now sold it at the new, lower price without it. CC & RS are the ones who are claiming they sold it without service, but didn't so mark all of the units.
BTW, my understanding is that most or all the units in the Radio Shack stores had stickers on them that said "Service Fee required for use" etc. Read the FW/etc posts (if you care to wade through 10000 messages) - people bought them anyways at $150 and figured they scored when they initially came up as "fee paid" - until RS told them to turn it off.
Disclaimer: My wife and I own two ReplayTV 5040's, one upgraded to 160GB, and they've totally changed our TV-watching habits. Once you use one, you won't go back. And their network capability is wonderful, as is DVArchive (which allows you to move shows to, or watch shows from, the PC's harddisk). DVArchive is also totally happy running on Linux, BTW.
My great-great-great(?) grandfather, "Madman" Henson, was one of the aviation pioneers of the mid-1800's. He designed a heavier-than-air plane, and flew models of it back around 1850 (1853?). The models were on the order of 15-20' wingspan I think. The full plane "ARIEL - The Henson Aerial Steam Carriage" was to have a wingspan of 150'. He was fully aware of Cayley, and probably knew him. Image
Eventually he gave up because steam engines just didn't have the power-to-weight ratio and moved on to other things, such as breech-loading-cannons (the family has a letter from the Dept of the Navy telling him, if I remember, that they were impractical/impossible).
He started his work in England, and moved to the US. His assistant, Stringfellow, continued making models and is fairly well-known in early aviation history. You can find a reproduction of Stringfellow's gliders in the Franklin Institute in Philladelphia, and last I knew the Smithsonian had either an original glider or a full-size reproduction.
When we went to the Smithsonian in the mid-70's to donate his papers, they took us into the closed section (renovations) to show us "Henson's glider". My mother said "that's not his glider, that's Stringfellow's" (we had most of his original drawings).
When I was, oh, 11 or 12 I was interviewed by phone by the London Sun about him. They must have gotten our names from the Smithsonian I imagine.
My old road had all the house numbers backwards for a 2 mile stretch or so on Mapquest and several other mapping services for a few years. The base problem (that they all shared) was that the US Census TIGER database had the house number direction backwards (for Skelp Level Rd in Downingtown, PA between Boot Rd and Harmony Hill Rd, to be exact).
This was doubly problematic because DSL companies used Mapquest and similar databases to get driving distances to the local CO, and used that for pre-qualifying homeowners. So instead of 8000 ft (measured with loopback tests multiple times) their databases showed us as being 12053 feet - and their limit was 12000 feet. Houses on side streets 200 feet away were ~8500 feet from the CO according to their databases.
Eventually after sending requests to fix the street to mapquest, yahoo, mapblast, etc, most if not all of them fixed it.
From that page (http://ed-thelen.org/comp-hist/TheCompMusRep/TCMR -V15.html):
One of the first things we did after the architecture was determined, was to put a red light on the end of a board to signal failure. Then field service didn't have to figure out what was wrong, but just take out the board and send it to the factory. Then we asked ourselves, "If field service isn't needed for fault detection, why are they needed on the customer site at all? Have the customer do it without a service call." This creates a new problem. The replacement has to be a fool proof insertion, without any special switches or an umbilical cord which might confuse the customer. In the final design, any board could be pulled out of a running machine and put in another one without anything happening.
Another problem was uncovered. How would we know what board to send to the customer for replacement? Could we depend on a secretary to pull out a bad board, read the model number, and accurately repeat it on the telephone? We thought that would be too much to ask. We added a feature that let the system read the slot location, the error state, the model number, revision level, and serial number of the bad board, finally throwing in a modem so that the computer could report the bad board directly to field service at Stratus. The electronic mail message to the Stratus computer reports what failed and all the details of the occurrence. The typical scenario is that the Stratus home office then calls up the customer and tells him that his machine has a failure. The customer doesn't know it until he's told. By then, the replacement board is on its way by Federal Express.
Agreed. A local winery (Stargazer) is solar-powered, including the owner's house. They use 5-10KWhr/day, 15KWhr on a cloudy bottling day. They produce circa 600 KWhr/month from 40ish panels, which cost $40K (though with tax credits and some local grants the outlay was $27K). They sell extra KW back to PECO for $.20/KWhr (which is more than you pay for a KWhr, BTW) - they're probably avoiding paying around $50-100/month, plus getting around the same amount - say total $150/month savings, or $1800/yr. (Note: this is probably over-estimating the savings, it may be closer to $100/month). (PECO charges $0.12-0.14/KWhr, quite high.)
At $27K cost, assuming 20yr lifetime, and interest on the $27K @ 5.5% (possible but a very low rate historically)q, that's around $185/month amortized. So _if_ you get 20 years out of it with no repairs, you wouldn't be _too_ worse off. Of course, if electricity rates were to rise substantially, let's say double, then you'd probably be slightly ahead of the game.
If the cost/W dropped to 1/10th the current (this article implies a circa 1/20th drop), then you'd be WELL ahead of the game - circa $20/month amortized. Even 1/5th current, without any subsidies would be cira $8k investment and amortized around $60-70/month - still a win at current prices.
So sub-$1/W starts to make it economical in high-cost-electricity areas to consider conversion. Sub-$.50 makes it quite worthwhile even assuming repairs will be needed over the 20 years or if it doesn't last as long.
I'd love to do this (I have a good SE exposure especially in winter), but I couldn't do more than put a bite in my costs - we use heat pumps, and have a 4000+ sq ft, 70+-year-old rambling and so-so insulated house, with 70+ windows, many of them large. In the winter we might use as much as 8-10,000 KWhr's per month. (Though we're also installing a woodstove for cold days - we have LOTS of trees, and a woodstove will be much better when the outside temp goes below where the HP's are efficient - around 20-30 degrees or so.)
XP is definitely better than say 98 in boot speed.
Back in '91 or so, an Amiga 3000 (25MHz 68030) could boot to gui (Workbench) and bring up TCP/IP & NFS and do NFS mounts in 7 seconds (warm boot; cold boot was ~12 seconds). And that was with 40MB 3600RPM slow-seeking drives, and a processor that was (even in raw MHz) 1/100 the speed of modern ones. Admittedly, it wasn't starting anywhere near as much, though it was bringing up the GUI, TCP, NFS, etc.
In 1984 our Stratus 200 fault-tolerant 'minicomputer' (68010-based) would let you yank a running CPU, and it would phone home to Stratus, they'd check the system info reported and if needed run remote diagnostics, and the phone would ring and they'd tell you to put the CPU back in (oh, and the system wouldn't even slow down).
Haven't read the patent, but sounds like prior art to me.
Sorry, the hard-buttons are pre-labelled, and while the LCD appears to be dot-matrix, I haven't seen any way to add graphics to the XML.
Check out the Harmony SST-768 at www.harmonyremote.com. They also have a new Harmony 659 with perhaps slightly better ergonomics for non-geeks, but not quite as powerful as the 768. (The 768 isn't bad for non-geeks, but the 659 has more pre-assigned buttons for things like "Watch TV" instead of having to select it on the LCD with the scroll wheel). Not to mention the built-in program guide (useful mostly if you don't have a sat/cable box with a guide, or if it's a slow, painful guide like the TVGuide guide).
Both of them will control PVR's, etc. The "Geek Eye" people were using the 768, which is overkill for that household - the 659 would be a better choice.
So it's not Linux... it rocks (so does Tivo, so no flames please). Both are wonderful for skipping commercials, backing up after "what did he say?", dealing with "Honey, can you come here for a second?", etc. Not to mention effortlessly make sure that you don't miss Crash on Junkyard Megawars....:-)
1) 30 second skip button without having to reprogram 2) Instant replay, skip forward/back N minutes, skip to minute N of the program 3) Can use all free disk space for buffering live TV 4) Networking - unlike Tivo, networking is built into the base unit. Stop playing on one, go to another room, resume playing there - way cool. Tell it to record something on the other Replay, easy. 5) DVArchive (open source java program) - make your Replay(s) think your PC is a Replay, and stream/download shows to it. Even better, the PC can serve them back to the Replays. And you can burn the files to DVD's or VideoCD's. 6) Trivial to add larger drives to. 7) Great IPG, and Replay Zones are great (preset searches/lists of programs, like SciFi movies)
ReplayTV (or Tivo) will change the way you deal with TV. No doubt about it.
These 'geeks' had trouble with the Harmony Remotes. It's not clear why, but I've found Harmony remotes are wonderful, and are great for the non-geeks in your life.
One of the very nice things compared to most multi-device remotes is that you don't have to spend time programming macros, worrying about which items need to be turned on or off or which input selected. You tell the website what devices you have, and which is connected to what, and it sets up "activities" like "Watch Television" or "Play a DVD" or "Listen to Radio". When you select one, the Harmony turns on or off all the boxes that need to be, and does any required switching automatically. No macros to program. It also changes the remote so that the buttons are all relevant to what you're doing (in "Play a DVD", the play button is the DVD play button, etc).
It's also a learning remote, so if you have truely oddball equipment like my Unity Motion big-dish satellite HD receiver, you can learn the codes in a few minutes.
And if you're a _true_ geek, you can reprogram it in XML.
(I had Mod points I could have used on this thread, but I couldn't resist posting.)
In addition to Compuserve and the like, PlayNet (the software for which later became AOL after it was ported to the PC by Quantum (now AOL)) had a very developed merchandise purchase system. Note that this was not per se over the internet; users dialed into Telenet or Tymnet X25 pads, and then connected to our servers via Telenet/Tymnet's network. You could view images, select colors, etc, and payment was via your CC account. This was all developed in the circa '85 timeframe.
Note: I was the person who coded the C64 side of this for PlayNet.
In 97 or so, there was a scammer on EBay who was fleecing the Japanese sword collector community. He knew something about swords, and had done some real deals, but then started ripping people off - both not sending things and cashing money orders (this was before paypal/etc), and offering to repair/polish blades, taking the swords, and not sending them back. He was in the midwest at the time. People started to figure it out, and got together. He made the big mistake of ripping off a Deputy elsewhere in the midwest as well.
He switched screen names and moved out of the town he lived in. He started scamming again under the new name, and I both identified him by his use of his real name to sign an email, and I proved he was using an image from someone else's website as the sword he was selling. We arranged for one of us (using a new screenname) to be the high bidder (the Deputy from KS). This gave us an address (Mailboxes/etc I think) in the northwest. He was arrested and forced to return about a dozen swords (and money I think), and I think was given a suspended sentence. He'd probably scammed on the order of $20-40K or more; one of the bigger ones at the time.
Note: while I helped track him down, I was lucky and wasn't taken by him, so after he was caught I only heard a few random details.
I have a paper written by some CMU students in 1988 titled: "Speculations on The Personal Computer of the Year 2000" from an Apple contest.
They did AMAZINGLY well for predictions in 1988:
Random predictions from the article (with comments):
PC's will eventually replace workstations. (check) Gallium arsenide will (GaAs) be common in consumer products (nope). Neural nets for pattern recognition (not really). Fiber-optic board/chip interconnects (nope). Parallel processing not uncommon, but coarse-grain (message passing, etc). (check)
Increased user-friendliness (yeah, easy): WSWIWG (Check). Templates for things like term papers, etc. (Check). In-line drawing in a document (check). Handwriting input (check).
Flatbed scanners common (check). Displays will be 2048x2048 (nope, but getting closer). Hundreds if not thousands of colors out of millions (underestimated). EL display will challenge CRT's and color EL's will be the only choice for laptops (Nope!) Video digitizers will be cheap and fast (yup). Considerable market for video applications (3d, animations, video-editing). (yup) Desktop video (Actually this was only a few years away when they predicted it: the Amiga). Visual programming (some). Speech recognition common and reliable (mostly). Sound digitizers cheap and common (yup). Braille pads (yes). Countour scanners as specialty items (yes). Natural language processing will be better though not excellent (not really).
Information infrastructure: High speed data networks locally, regionally, nationally. (Yup!) Information explosion (absolutely). Individuals will start storing data on the network and making it available to others. (And how!) Fiber-optic will be used within buildings or in neighborhoods (neighborhoods yes (cable HFC), in buildings generally no). Cellular telephone will be integrated with laptops (sort of - 802.11 with hotspots, ricochet, etc is close). Can access your computer from elsewhere in the net (Oh yes). Mass proliferation of data sources. (beyond all expectations) BBS's may morph into huge multimedia multi-user discussion boards (like Slashdot...) Common Information Protocol (think HTTP, XML, etc - absolutely). Hypermedia (yes, though more ubiquitously than they expected). Information Filters (coming, but needs more work). Computers will become available in libraries (yes). Telecommuting, and shift towards white-collar workers. (yup) Ubiquitous electronic mail (yes).
Storage: CD's will become like floppies (Absolutely). 200 MB to 1GB on a harddisk (rather low; more like 1996 - still good guess for 1988). HD's will be the size of a pack of cards. (check) The small ones like that will store 40MB (WAY low).
It will be more difficult to deceive the public. (Uhhh, maybe. Yes.)
Applications: Simulations and models will be important. (Yes, mostly). Visual design of things like architecture. (Yes). Serious computer art (yes). Quick but costly 3d prototyping (yup). Auto-transcription of music (yes). 3d-realtime views of patients internals (ala NMR aka MRI) (not realtime, but easy/cheap)
(my fingers get tired)
Heavy use of object orientation (yes). 16Mb dram (way under). 64-bit CPU's common (nope, but coming) 25 MIPS processors (ROTFL!!!!!) Base 32MB of memory on motherboard (well under). Cheap color laserprinters (almost). Cheap quiet color inkjets (yup). dot matrix fading (gone).
Slashdot Mirror
Note the site says explicitly that they're in "early stages of development", and one thing they hope to achieve is real-time decode. Note they don't even mention real-time encode, which is probably an order of magnitude harder.
It does sound interesting, however as people here have noted a number of techniques it apparently uses (such as arithmetic coding, probably wavelets in general, etc) are known to be patented. H.264 Baseline Profile doesn't use arithmetic coding (CABAC) because Baseline is designed to avoid most/all of the patent-licensing issues. H.264 Main Profile uses CABAC, etc (and gets better compression because of it).
Back in the day (Amiga 3000 introduction circa 1990) we could boot an Amiga off a 40MB disk from power-on (including loading a replacement for the ROM image) to fully up (including mounting NFS mounts) in ~15 seconds; 7 seconds from a warm boot. Effectively we chunked the load of much of the OS (including GFX, Workbench (desktop), etc) as part of the softload of the ROM image into RAM. On top of that, on warm boot we checksummed the ROM image in RAM (as well as various major OS library structures), and if they were good we didn't reload them. (Remember, the Amiga ran without memory protection.) Some of the boot script that followed would cause overlap (anything tha spawned servers/processes), other parts were sequential but generally were very fast and often cached (like "ASSIGN XYZ: dh0:xyz" commands and the like).
At the time, Win3.1 took Quite A While to boot. Getting past the BIOS alone usually took longer than the Amiga took to finish booting.
Trying to do all the reordering in the OS (as suggested in several posts here) seems like a good idea, but ignores some issues:
Tagged queuing in SATA/etc drives is a step in the right direction, though last I checked it wasn't equal to TQ in SCSI. Native Command Queuing in SATA will probably give similar performance to TQ in SCSI.
This PDF on Native Command Queuing is even more interesting.
ATA is definitely not SCSI-over-IDE.
:-)
ATAPI is SCSI-over-IDE however.
I wrote the IDE/ATA drivers for the Amiga. The Amiga SCSI drivers accepted "SCSIDirect" commands from applications. Internally, all IO commands were converted to SCSIDirect commands for execution. To implement ATA, I added a SCSIDirect->ATA translator (which wasn't that hard - about 3 weeks from start to working, booting system - and I implemented just about all SCSI commands even semi-reasonable (all of CCS I think, plus quite a bit).
Doing it this way made implementing support for ATAPI CDROMs (something I did as a contract after Commodore folded) Very Easy.
I'm fairly certain (having been on the Amiga OS team at the time, and having worked with Bill a fair bit) that he was not stiffed by Commodore on anything. I don't know what sort of deal there was for including ARexx in the OS - it might not have included money.
Bill did a fair bit of work for Commodore on contract, and also did a lot of testing and tool development for us (evil memory-allocation failure tools, for example). We tried to hire him on multiple occasions, but he preferred to stay in Boston.
Bill seems to be involved in Linux kernel/etc stuff nowadays from a quick google search.
Arexx.... yum (I was on the Commodore Amiga OS team at the time).
BTW, Gnus in Emacs also _totally_ integrates mail and news. Plus like all things emacs, it's totally configurable/modifiable. Though elisp is much more of a pain to program in the ARexx.
I know the author (college), and I can tell you he's on a number of mailing lists, including the extended-college-ilk list. Not to mention that he's a TV star... ;-) You've probably watched him (and rooted for him) on TV if you're a slashdotter/techie.
The final blow was when in fall of '93 Mehdi decided to build a few 10's of thousands of the new (AA/AGA) machines (A1200, etc), and 300,000+ of the old chipset-based machines (A600). Needless to say, the old machines didn't move off the shelves very fast at Xmas, and that was the final nail.
There were other instances like that too. Mostly it was caused by not following up on successful products (C64, A500, to some extent A3000) and trying to milk them for too long. The A1200 was the right machine; it was just too late by a year or two. Engineering had it's issues too, in particular biting off more than we could chew on the total redesign of the chipset which was never quite finished ("AAA"), and not giving enough attention to the potential high-volume products, though in general engineering was pretty focused on them.
It's tough when the CEO won't let marketing talk to engineering directly, and insists all contact go through him and his cronies... Disclaimer: I'm an ex-Commodore engineer from these times, and after bankruptcy was declared, we burnt Mehdi Ali in effigy in my backyard (literally).
The retailer bought the units from Replay with the 3-year-activation for $XXX, and was reselling them for ~$500.
When Replay decided to change their model, the retailer took the option of selling the units without the activation for $150, and getting a check back from Replay for circa $200 (on their old stock). When they made that deal, it was incumbent on them to sell them without activation in order to get the check from Replay. They (CC) sold some of them (not all) without telling the consumer that the activation had been removed, but they then went and told Replay (or had already told Replay) that those units had been sold without activation.
It doesn't matter what's sealed in the box if the retailer (correctly) tells you that the deal is different. The retailer changed the deal, but didn't tell (all) the consumers. Replay has no direct knowledge of which consumers the retailer sold to under the old model or the new model, it only knows that CC told Replay that all those units (at $150) were sold without activation.
Note again that some CC's correctly marked them, and others marked them sometime during the day. Also note that most Radio Shacks correctly marked them _and_ the boxes had stickers that said activation required.
Circuit City screwed up, bigtime. One way or another, they're liable (though perhaps only to refund purchasers their money, perhaps to Replay (i.e. circuit city doesn't get the $250ish from Replay, and Replay provides the service that CC sold them)).
CC will probably claim it can't determine which people were told and which weren't, and so will probably only offer to refund people's $150ish in exchange for the units back.
The seller is not a 3rd party to the sale, they're a 1st party. They are not merely an agent for Replay; they buy inventory from Replay, then sell that to consumers. Replay can't tell them to mark the stock down and remove the activation, and Replay can't go in and demand they change the stickers. Replay can make a deal with the retailer, though, for the retailer to remove the activation and get paid by Replay when they sell it or when the user activates it, etc. In that case, it's incumbent on the retailer to live up to that bargain, and so inform the buyers of what they're buying.
Remember, when you buy something in a store you're buying it from the retailer. The manufacturer has a requirement to honor the deal they sold it to the retailer under - but in this case the retailer and the manufacturer agreed to a new deal that involved the retailer remarking the unsold stock, and the retailer screwed it up.
The phone reps at Replay couldn't know that CC has screwed up, nor that CC had jumped the gun and dropped the price early.
You stated that Replay "did NOT tell them to take off the markings". There is considerable evidence that they did, but that CC didn't do a good job of getting the website people and all the store managers involved in getting word out and remarking the shelves/stock.
This story was in the NYTimes Dec 18th: NYTimes article - costs money to read now in their Circuits section.
Fundamentally, the purchase and promise of service was by the retailer. They separately had purchased the units (originally with a promise of service) from Replay.
Replay had told them it was changing the pricing model, and so that the retailers didn't end up holding the bag, they told the retailers they could sell the units without service and Replay would refund to them most or all of the service contract price. Part of that deal was that they were to be sold without the service included (and mark the units clearly as such), and that the retailers would report back which units were sold without service. (I know none of this directly, but I believe the details are very close to what I've described.)
CC sold the units, and in many stores (and on the web) didn't tell the buyers that these were now without service (which was why there was such a discount, since CC was to get a check back from Replay for each unit sold). CC is liable in some manner: either to the customer to provide service (they shouldn't have told Replay they sold it without service when they sold it with service), to the customer for a full refund (to correct for the mistake), or they're liable to Replay (Replay ends up providing service anyways, but doesn't pay CC).
Consumers who bought units at Radio Shack generally are out of luck (unless an employee told them it came with service - a few did, but many of them apparently knew to tell people service was extra, and the boxes were apparently mostly marked "service activation required" etc). Those people were really gambling that Radio Shack and Replay had simply made a mistake and they'd slip through unnoticed.
IMHO, IANAL, etc.
Or to put it another way, for all that sit salivating for the next "oops" rebate/etc snafu:
TANSTAAFL
(look it up if you don't get it)
Replay has also been adding features, such as one to mostly replace (and in some way surpass) commercial-skip - Replay 5000's/5500's can now skip to the next or previous show segment (which may conveniently skip over all the commercials without having to hit the 30-sec skip key 6 times, then back up a few times).
It's also handy for skipping to the next segment of a show when they're doing a report/etc you don't care about.
As for your legal comments - it's not clear here that those comments apply to Replay or to the retailer. Or, if you prefer, the retailers are in violation for telling replay that the consumers bought the product without the 3-year service, when the retailer failed to remove the 3-year-service stickers. (Note that almost all Radio Shack units did NOT have 3-year-service stickers, though some RS employees told people it was included, and that some CC's (especially by later in the day) had removed/X-ed-out the stickers and had signs up).
First of all, the beef would be with the seller, not directly with Replay. And they can simply refund your money and take it back; that's generally considered complete restitution in a case such as this. Even under your math, Replay doesn't owe anyone $616 or $449; at most (if your interpretation were correct) was 3 years of service. The cost of that service (if bought) is $299; of course it costs them less, but that service fee subsidizes the below-cost-of-goods sale prices that both TiVo and Replay are now selling at.
As others may have said, Replay told retailers to remove the 3-year stickers, etc before dropping the price on all current inventory. CC and Radio Shack didn't bother to do this, and didn't update their websites either. Replay had set the price change for around Dec 20th; CC and Radio Shack (to get extra Xmas sales I assume) jumped the gun and did it about a week early. That's not Replay's fault. And the reps called to ask about the price answered correctly so far as they knew - the retailers were supposed to wait to change the price and the terms.
All those FatWallet-ers who bought the units knew it was an unintentional mistake (and probably knew it was the retailer's mistake). They figured they were getting something for nothing (effectively).
Basically, Replay is getting screwed over here by CC and the Shack (who also screwed themselves). Many of those units (bought by FW-ers) will be returned, and have to be sold eventually as refurbs or open-box units. (Earlier this year, Replay sold refurbs with lifetime service for ~$300 (service was $250 then)).
CC and RS are telling Replay "cancel the activation on these serial numbers, we sold them without service, then send us a check" since CC/RS paid for that inventory when it did include service, but they now sold it at the new, lower price without it. CC & RS are the ones who are claiming they sold it without service, but didn't so mark all of the units.
BTW, my understanding is that most or all the units in the Radio Shack stores had stickers on them that said "Service Fee required for use" etc. Read the FW/etc posts (if you care to wade through 10000 messages) - people bought them anyways at $150 and figured they scored when they initially came up as "fee paid" - until RS told them to turn it off.
Disclaimer: My wife and I own two ReplayTV 5040's, one upgraded to 160GB, and they've totally changed our TV-watching habits. Once you use one, you won't go back. And their network capability is wonderful, as is DVArchive (which allows you to move shows to, or watch shows from, the PC's harddisk). DVArchive is also totally happy running on Linux, BTW.
Image
Eventually he gave up because steam engines just didn't have the power-to-weight ratio and moved on to other things, such as breech-loading-cannons (the family has a letter from the Dept of the Navy telling him, if I remember, that they were impractical/impossible).
He started his work in England, and moved to the US. His assistant, Stringfellow, continued making models and is fairly well-known in early aviation history. You can find a reproduction of Stringfellow's gliders in the Franklin Institute in Philladelphia, and last I knew the Smithsonian had either an original glider or a full-size reproduction.
When we went to the Smithsonian in the mid-70's to donate his papers, they took us into the closed section (renovations) to show us "Henson's glider". My mother said "that's not his glider, that's Stringfellow's" (we had most of his original drawings).
When I was, oh, 11 or 12 I was interviewed by phone by the London Sun about him. They must have gotten our names from the Smithsonian I imagine.
William Samuel Henson"
My old road had all the house numbers backwards for a 2 mile stretch or so on Mapquest and several other mapping services for a few years. The base problem (that they all shared) was that the US Census TIGER database had the house number direction backwards (for Skelp Level Rd in Downingtown, PA between Boot Rd and Harmony Hill Rd, to be exact).
This was doubly problematic because DSL companies used Mapquest and similar databases to get driving distances to the local CO, and used that for pre-qualifying homeowners. So instead of 8000 ft (measured with loopback tests multiple times) their databases showed us as being 12053 feet - and their limit was 12000 feet. Houses on side streets 200 feet away were ~8500 feet from the CO according to their databases.
Eventually after sending requests to fix the street to mapquest, yahoo, mapblast, etc, most if not all of them fixed it.
From that page (http://ed-thelen.org/comp-hist/TheCompMusRep/TCMR -V15.html):
One of the first things we did after the architecture was determined, was to put a red light on the end of a board to signal failure. Then field service didn't have to figure out what was wrong, but just take out the board and send it to the factory. Then we asked ourselves, "If field service isn't needed for fault detection, why are they needed on the customer site at all? Have the customer do it without a service call." This creates a new problem. The replacement has to be a fool proof insertion, without any special switches or an umbilical cord which might confuse the customer. In the final design, any board could be pulled out of a running machine and put in another one without anything happening.
Another problem was uncovered. How would we know what board to send to the customer for replacement? Could we depend on a secretary to pull out a bad board, read the model number, and accurately repeat it on the telephone? We thought that would be too much to ask. We added a feature that let the system read the slot location, the error state, the model number, revision level, and serial number of the bad board, finally throwing in a modem so that the computer could report the bad board directly to field service at Stratus. The electronic mail message to the Stratus computer reports what failed and all the details of the occurrence. The typical scenario is that the Stratus home office then calls up the customer and tells him that his machine has a failure. The customer doesn't know it until he's told. By then, the replacement board is on its way by Federal Express.
Agreed. A local winery (Stargazer) is solar-powered, including the owner's house. They use 5-10KWhr/day, 15KWhr on a cloudy bottling day. They produce circa 600 KWhr/month from 40ish panels, which cost $40K (though with tax credits and some local grants the outlay was $27K). They sell extra KW back to PECO for $.20/KWhr (which is more than you pay for a KWhr, BTW) - they're probably avoiding paying around $50-100/month, plus getting around the same amount - say total $150/month savings, or $1800/yr. (Note: this is probably over-estimating the savings, it may be closer to $100/month). (PECO charges $0.12-0.14/KWhr, quite high.)
At $27K cost, assuming 20yr lifetime, and interest on the $27K @ 5.5% (possible but a very low rate historically)q, that's around $185/month amortized. So _if_ you get 20 years out of it with no repairs, you wouldn't be _too_ worse off. Of course, if electricity rates were to rise substantially, let's say double, then you'd probably be slightly ahead of the game.
If the cost/W dropped to 1/10th the current (this article implies a circa 1/20th drop), then you'd be WELL ahead of the game - circa $20/month amortized. Even 1/5th current, without any subsidies would be cira $8k investment and amortized around $60-70/month - still a win at current prices.
So sub-$1/W starts to make it economical in high-cost-electricity areas to consider conversion. Sub-$.50 makes it quite worthwhile even assuming repairs will be needed over the 20 years or if it doesn't last as long.
I'd love to do this (I have a good SE exposure especially in winter), but I couldn't do more than put a bite in my costs - we use heat pumps, and have a 4000+ sq ft, 70+-year-old rambling and so-so insulated house, with 70+ windows, many of them large. In the winter we might use as much as 8-10,000 KWhr's per month. (Though we're also installing a woodstove for cold days - we have LOTS of trees, and a woodstove will be much better when the outside temp goes below where the HP's are efficient - around 20-30 degrees or so.)
XP is definitely better than say 98 in boot speed.
Back in '91 or so, an Amiga 3000 (25MHz 68030) could boot to gui (Workbench) and bring up TCP/IP & NFS and do NFS mounts in 7 seconds (warm boot; cold boot was ~12 seconds). And that was with 40MB 3600RPM slow-seeking drives, and a processor that was (even in raw MHz) 1/100 the speed of modern ones. Admittedly, it wasn't starting anywhere near as much, though it was bringing up the GUI, TCP, NFS, etc.
In 1984 our Stratus 200 fault-tolerant 'minicomputer' (68010-based) would let you yank a running CPU, and it would phone home to Stratus, they'd check the system info reported and if needed run remote diagnostics, and the phone would ring and they'd tell you to put the CPU back in (oh, and the system wouldn't even slow down).
Haven't read the patent, but sounds like prior art to me.
Check out the Harmony SST-768 at www.harmonyremote.com. They also have a new Harmony 659 with perhaps slightly better ergonomics for non-geeks, but not quite as powerful as the 768. (The 768 isn't bad for non-geeks, but the 659 has more pre-assigned buttons for things like "Watch TV" instead of having to select it on the LCD with the scroll wheel). Not to mention the built-in program guide (useful mostly if you don't have a sat/cable box with a guide, or if it's a slow, painful guide like the TVGuide guide).
Both of them will control PVR's, etc. The "Geek Eye" people were using the 768, which is overkill for that household - the 659 would be a better choice.
So it's not Linux... it rocks (so does Tivo, so no flames please). Both are wonderful for skipping commercials, backing up after "what did he say?", dealing with "Honey, can you come here for a second?", etc. Not to mention effortlessly make sure that you don't miss Crash on Junkyard Megawars.... :-)
1) 30 second skip button without having to reprogram
2) Instant replay, skip forward/back N minutes, skip to minute N of the program
3) Can use all free disk space for buffering live TV
4) Networking - unlike Tivo, networking is built into the base unit. Stop playing on one, go to another room, resume playing there - way cool. Tell it to record something on the other Replay, easy.
5) DVArchive (open source java program) - make your Replay(s) think your PC is a Replay, and stream/download shows to it. Even better, the PC can serve them back to the Replays. And you can burn the files to DVD's or VideoCD's.
6) Trivial to add larger drives to.
7) Great IPG, and Replay Zones are great (preset searches/lists of programs, like SciFi movies)
ReplayTV (or Tivo) will change the way you deal with TV. No doubt about it.
These 'geeks' had trouble with the Harmony Remotes. It's not clear why, but I've found Harmony remotes are wonderful, and are great for the non-geeks in your life.
One of the very nice things compared to most multi-device remotes is that you don't have to spend time programming macros, worrying about which items need to be turned on or off or which input selected. You tell the website what devices you have, and which is connected to what, and it sets up "activities" like "Watch Television" or "Play a DVD" or "Listen to Radio". When you select one, the Harmony turns on or off all the boxes that need to be, and does any required switching automatically. No macros to program. It also changes the remote so that the buttons are all relevant to what you're doing (in "Play a DVD", the play button is the DVD play button, etc).
It's also a learning remote, so if you have truely oddball equipment like my Unity Motion big-dish satellite HD receiver, you can learn the codes in a few minutes.
And if you're a _true_ geek, you can reprogram it in XML.
(I had Mod points I could have used on this thread, but I couldn't resist posting.)
In addition to Compuserve and the like, PlayNet (the software for which later became AOL after it was ported to the PC by Quantum (now AOL)) had a very developed merchandise purchase system. Note that this was not per se over the internet; users dialed into Telenet or Tymnet X25 pads, and then connected to our servers via Telenet/Tymnet's network. You could view images, select colors, etc, and payment was via your CC account. This was all developed in the circa '85 timeframe.
Note: I was the person who coded the C64 side of this for PlayNet.
In 97 or so, there was a scammer on EBay who was fleecing the Japanese sword collector community. He knew something about swords, and had done some real deals, but then started ripping people off - both not sending things and cashing money orders (this was before paypal/etc), and offering to repair/polish blades, taking the swords, and not sending them back. He was in the midwest at the time. People started to figure it out, and got together. He made the big mistake of ripping off a Deputy elsewhere in the midwest as well.
He switched screen names and moved out of the town he lived in. He started scamming again under the new name, and I both identified him by his use of his real name to sign an email, and I proved he was using an image from someone else's website as the sword he was selling. We arranged for one of us (using a new screenname) to be the high bidder (the Deputy from KS). This gave us an address (Mailboxes/etc I think) in the northwest. He was arrested and forced to return about a dozen swords (and money I think), and I think was given a suspended sentence. He'd probably scammed on the order of $20-40K or more; one of the bigger ones at the time.
Note: while I helped track him down, I was lucky and wasn't taken by him, so after he was caught I only heard a few random details.
I have a paper written by some CMU students in 1988 titled: "Speculations on The Personal Computer of the Year 2000" from an Apple contest.
They did AMAZINGLY well for predictions in 1988:
Random predictions from the article (with comments):
PC's will eventually replace workstations. (check)
Gallium arsenide will (GaAs) be common in consumer products (nope).
Neural nets for pattern recognition (not really).
Fiber-optic board/chip interconnects (nope).
Parallel processing not uncommon, but coarse-grain (message passing, etc). (check)
Increased user-friendliness (yeah, easy):
WSWIWG (Check).
Templates for things like term papers, etc. (Check).
In-line drawing in a document (check).
Handwriting input (check).
Flatbed scanners common (check).
Displays will be 2048x2048 (nope, but getting closer).
Hundreds if not thousands of colors out of millions (underestimated).
EL display will challenge CRT's and color EL's will be the only choice for laptops (Nope!)
Video digitizers will be cheap and fast (yup).
Considerable market for video applications (3d, animations, video-editing). (yup)
Desktop video (Actually this was only a few years away when they predicted it: the Amiga).
Visual programming (some).
Speech recognition common and reliable (mostly).
Sound digitizers cheap and common (yup).
Braille pads (yes).
Countour scanners as specialty items (yes).
Natural language processing will be better though not excellent (not really).
Information infrastructure:
High speed data networks locally, regionally, nationally. (Yup!)
Information explosion (absolutely).
Individuals will start storing data on the network and making it available to others. (And how!)
Fiber-optic will be used within buildings or in neighborhoods (neighborhoods yes (cable HFC), in buildings generally no).
Cellular telephone will be integrated with laptops (sort of - 802.11 with hotspots, ricochet, etc is close).
Can access your computer from elsewhere in the net (Oh yes).
Mass proliferation of data sources. (beyond all expectations)
BBS's may morph into huge multimedia multi-user discussion boards (like Slashdot...)
Common Information Protocol (think HTTP, XML, etc - absolutely).
Hypermedia (yes, though more ubiquitously than they expected).
Information Filters (coming, but needs more work).
Computers will become available in libraries (yes).
Telecommuting, and shift towards white-collar workers. (yup)
Ubiquitous electronic mail (yes).
Storage:
CD's will become like floppies (Absolutely).
200 MB to 1GB on a harddisk (rather low; more like 1996 - still good guess for 1988).
HD's will be the size of a pack of cards. (check) The small ones like that will store 40MB (WAY low).
It will be more difficult to deceive the public. (Uhhh, maybe. Yes.)
Applications:
Simulations and models will be important. (Yes, mostly).
Visual design of things like architecture. (Yes).
Serious computer art (yes).
Quick but costly 3d prototyping (yup).
Auto-transcription of music (yes).
3d-realtime views of patients internals (ala NMR aka MRI) (not realtime, but easy/cheap)
(my fingers get tired)
Heavy use of object orientation (yes).
16Mb dram (way under).
64-bit CPU's common (nope, but coming)
25 MIPS processors (ROTFL!!!!!)
Base 32MB of memory on motherboard (well under).
Cheap color laserprinters (almost).
Cheap quiet color inkjets (yup).
dot matrix fading (gone).