The linked article is awful. You will not understand PenTiles from it. Go to the source to get the facts.
Briefly...
Human perception has lower spatial resolution for blue, so have fewer blue pixles. this has nothing to do with intensity sensitivity.
Apparently the column drivers on an LCD cost more than the row drivers. I have no idea why, but I will accept that. The pentiles use twice as many row drivers for red and green to boost the spatial resolution without requiring more column drivers.
Much like cleartype, they can position pixels on other than the natural boundaries to accomplish subpixel effects. The example on the page is a special case of a single white point. Mostly this will be useful for smoothing edges.
It is a really well written desription, it is a shame Design Engineering didn't have an writer that could understand it.
The clothing and laundry machine industries are already gearing up to put rfid tags in garments. The idea is that your washer and dryer will watch what you put in them and then behave accordingly or warn you that you are about to turn all your underwear pink.
Oh look, I went searching for a link, and even found one on slashdot
Years ago the informatics crowd I hung with was working on the relationship between computation and power.
It turns out that because you need to represent state in a stable mechanism, and you can't change the stable state without using energy that there is a lower bound on the energy required to perform a given calculation.
I have forgetten everything else, and I suspect it is many orders of magnitude smaller than current cpu power usage, but it is enough to break the '==' operator on the parent's title.
Sapping heat from the smokestack contents will probably cause it to not work correctly.
The goal of a smokestack is to get the harmful exhaust away from the ground long enough that it disperses sufficiently before touching down.
This is done with convection. The hot gas in the tall stack creates the draw that powers it and blows the plume up after it leaves the stack, the hot plume continues to lift itself until it bleeds off too much heat, then it starts coming back down, but presumably dispersed enough to not be too noxious.
The smoke stack was designed with a known gas temperature and dispersal requirement and a desire to minimize masonry. If you take away heat from the gas you will reduce your plume altitude and cause it to come down in a more concentrated region.
I doubt you can use the thermo-generated electricty to run blowers to compensate. The `no free lunch' law of thermodynamics will probably forbid that. (Unless blowers are much more efficient than convection.)
Now, if you are just bleeding off waste steam then it would work, but most of the energy in steam is the expansion from water to steam, there is relatively little left in the puffy clouds.
Mostly unrelated note: I used to live in Pittsburg in a community where all the houses were required to have slate roofs, stone or brick exteriors and no wood trim. Even the window frames were metal. It was a fire-proof community from the days when the steel mills spewed lots of solids including hot cinders. The plume was powerful enough to carry those large distances fast enough that they were still hot enough to start a fire.
If you have a decent sound card (not the eMachine I am using) then the digital-analog-digital damage is going to be far less than the mp3 or ogg encoding will do.
Just write yourself a little program to...
wait a second
start recording and start track X off the cd
when the track ends stop recording
trim the silence off the ends of the track
encode the mp3
repeat for all tracks
Ok, you will be ripping real time, but big deal. Let it go overnight. You will also need to type in your own track info until someone writes a new freedb-like service that uses a fuzzy audio signature instead of the digital signatures.
No special hardware or loopback cables are required. (well, maybe one cable if your machine doesn't let you route CD audio to the DAC input) Just a different ripper than you are used to.
extra cost for powered etherenet devices or the electrical work to power
Wasn't clear to you? It might help to read the acual[sic] comment once in a while.
:-)
I'll be more clear. Basically no one has powered ethernet in their closets today, so you either have to buy the 3com power adding device mentioned in the article or replace all your ports with power supplying ports. For new build outs I presuming that powered port devices are more expensive than non-powered.
Ok, you want a hub. Go down to compusa and grab a 5 port 10/100 hub for $50. Save the $150 + install and if someone steals your hub buy another one.
I don't see much point in using a switch in your office, unless you really do a lot of high speed transfer between your cubicle machines or have an awful network. If you REALLY need a switch, then that will set you back $60 instead.
I wonder where Tom got the numbers for the cost comparison? He has cable pulling at $300+ for each run, I pay about $120. If his cable puller is charging him quad for four wires he needs a new cable puller. It ain't rocket science. He has upstream switch ports at $90ea! I'm paying $25 each. Then he has the mystical `gigabit switch port' row.
The maintanence/year row implies that the maintanence cost of the 3com device is zero. I would rather have something like "15% of cost"/year for it. I don't have many 4 year old hubs that haven't either died or got some blown ports. Lots of lightning around here. Its especially hard on hubs and NICs because of the ground surge differential on close strikes.
I see no indication of either the extra cost for powered ethernet devices or the electrical work to power the 3com devices.
The traditional wiring is costed for the worst case, where 4 ports are really active. Needing 4 ports available everywhere is not the same as needing 4 ports active everywhere. I routinely run four times the copper that I will need and activate ports as needed.
An even better comparison would have been to compare the 3com wall jack hub to a free standing hub. But then the 3com would not have looked so good.
It's a shame 3com didn't have these during the dotcom boom. They could have sold dumptruck loads of them. Now? I doubt it.
Exactly the problem that popped up in my mind when I read it. I re-read and came to this paragraph in the Idiot Savants' Guide to Rubberhose...
Rubberhose relies on internal maps to locate where the actual bits of your data are stored amid the random characters. Each aspect has its own corresponding map, and you can only decrypt that aspect's map when you type in the passphrase for that particular aspect. This is why you need to type in all the passphrases before you can safely write to a Rubberhose disk.
Re:Stop using stupid C language - good point
on
Wu-ftpd Remote Root Hole
·
· Score: 4, Insightful
It is a good point. Poorly made, but there is a good point hiding in there. I see the article has atracted 6 flame replies and a -1 troll before I read it.
I have not made an ontology, but it seems to me that nearly all exploits the past few years have been (in decreasing prevalence order)
data buffer overflow
string overflow
filename.. abuse
A language with safe memory management will eliminate the first two. The third needs a more robust set of filename functions.
Its not impossible, or even hard, to avoid these sins in C programming. But, it also isn't impossible or even hard to screw up and commit this sins.
Programmers make mistakes. That is why it is called programming instead of typing. Choosing a language that minimizes the security impact of mistakes makes a lot of sense.
Don't forget about other criteria. You may need the speed that can be had with well written C code. Usually you won't.
I look at my servers. They are all the slowest rackmount machines I could buy from Gateway when I bought them, 800MHz PIII is typical. (They are plural because the have different security policies, not because of load.) They handle things like mail, http, samba, cvs, ldap, the usual suspects for a 100 engineer software firm. They rarely go beyond 5% cpu utilization. I would gladly sacrifice my surplus cpu cycles for slower, safer, services. When they do go beyond 5% it is almost always for a very specific function like the rsync algorithms or blowing backup data over to another box. Make the hot spots of these functions fast, spend a lot of time making them secure. Probably not more than 400 lines of code between them. Let the rest be written in a safe language.
One other factor to consider is that the gzip transfer encoding compresses much better than the algorithm in the modem. Part of this is the algorithm with its larger dictionary size, the other part is the `pure' data stream being fed to it. It is just the html, not the html interspersed with ppp, ip, and tcp headers.
Speaking as the maintainer of a site that is periodically slashdotted...
Yes, a throttling server is a great idea. If you recognize that there will always be a load too high for you to handle (10 requests per minute for my site, yes minute, it is a physical device), then you must either decide to deal with the load or let the load crush your machine.
Consider a typical web server. When it gets overloaded it slows down, each request takes longer to handle, there are more concurrent threads, overall efficiency drops, each request takes longer to handle.... welcome to the death spiral. (on my site-which-must-not-be-named-less-it-be-slashdotte d, everyone waiting in queue gets a periodic update, at a certain point the load of generating the updates swamps the machine. I have to limit the number of people in queue.)
The key decision is to determine how many concurrent threads you can handle without sacrificing efficiency and then reject enough traffic to stay under that limit.
This is where optimism comes in and bites you in the ass. You remember that every shunned connection is going to cost you money/fame/clicks whatever so you set the limit too high and melt down anyway.
It is 15 minutes after the article post and the site is dead. I got to the part about calculating how much RAM was required per visitor and multiplying by the expected number of visitors.
Maybe they need to adjust their constants.:-)
It is those d*mn modem users that drive up the RAM use. They stay connected longer on their GET and tie up resources longer.
You are getting more info from your iPod than I am or the guy at neuron.com. Notice his kernel messages show an unknown type device and sequential ascii letters in all the text fields. This is the same thing I get. You on the other hand have it actually recognized as a SCSI disk!
I'll have to play around some more. I suspect my ohci1394 is goofed. Maybe I'll put debian on an imac so I can trace through that and see what is different.
I tried. I have firewire drives on my linux box. The iPod does not work as a disk `out of the box'. When the driver goes to read the `config' page it gets back garbage. Maybe there is a special command to flip the device into disk mode, maybe there is a bug in my linux 1394 stack, but it sure doesn't mount as a disk.
If I were smarter about OS X, I would figure out how to trace 1394 packets and just reverse engineer what happens on the Mac, then make that happen on Linux.
Damn you and your insightful pragmatism! You rob us of our micro-turbine powered dreams.
Fuel cells will win in efficiency. Probably by a large margin.
Perhaps turbines have other advantages...
cost - turbines could be very cheap. Micro fabrication can be cheap (not yet, later) and fuel cells require expensive catylsts. These might be well suited to disposable items.
power density - it looks like 20W of turbine will be much smaller and lighter than 20W of fuel cell. For many applications this difference will be dwarfed by the mass and volume of the fuel. Other applications, like short-mission robots will benefit from the lighter generator.
20W won't get your a dual athlon. I measured my Celeron 700 machine at 70W and a 1.1GHz Athlon box at 110W. Maybe 130-150W for a dual Athlon. You will need a little bank of 8 turbines to power the beast.
And about flashlight life... cockeyed has a series on "how much is inside". Checkout the battery one for a shocker about how much is really in a pair of D cell batteries. Then try to figure out why your flashlight is always almost dead. Who comes in and uses up your flashlight?
"But where does the exhaust go?" - well, out, that is rather the definition of exhaust. The exhaust is water vapor, unused combustion air, and heat. That shouldn't be a problem. Well, you won't want 20W to 40W of heat running in your pocket, but other than that it should be fine.
"And isn't this kind of a step back in our attempt to stop using fossil fuels..." - I am an American. My goverment has no such policy. All your oil are belong to us. For those of you in conservation minded countries, hyrdrogen is not a fossil fuel. It is a theoretically handy way to store electrical energy. There are technical hurdles. Not insurmountable, just insurmounted. Once there is a demand, there will be a way of distributing and storing the hydrogen.
Which Apple partition destroying software would that be? I must have missed that one. I am only aware of two.
The iTunes partition destroyer was pulled in something like 24 hours and replaced not long after.
Some years ago there was a problem with certain models of hard drives (Quantums I believe) that didn't handle their write caches well on a scsi reset. That went on for quite a while, but was not an issue with supported Apple hardware, it was some 3rd party drives that had tweaks to enable write behind caching. (The very large Oracle installation on Alphas that I work with had the same problem with them. Unable to resolve it with the vendors we finally scrapped all the disks and replaced them with a different vendor's drives.)
Got one already! Not only cool, but its the only shirt in my closet that my girls won't steal. Its "too scary".
Re:Problems with induction - hmm
on
Concept PC 2001
·
· Score: 2
It seems we live in worlds too different to even communicate. Allow me to supply more detail...
Zip - it never entered my mind to use a zip disk. I have no use for cheap, unreliable storge (floppies), I certainly have no use for expensive unreliable storage.
Bootable CDs - my prefered way to install software. Fragility doesn't matter. I lose them or loan them. I just burn a new one each time I need one.
Bootable Network - I use this. If my NIC dies I replace it. I'm using Linux on these machines. I do not have a driver issue to deal with. If I did, I would just use a CD or CD-R to tote the driver over.
Everything is networked. All my valuable data moves around at night using rsync so it is stored on multiple machines. rsync is patient. Most users (image editors aside) don't make enough data that rsync won't catch up in a couple of hours at night, even on a modem.
You use "floppies" and "reliable" in the same sentence. I'll assume that is a typo.:-)
You mention me using a costly horrible to upgrade box. I'm not sure what you mean there. I buy my wintel boxes for $400 (cheapest machine at BestBuy) and never upgrade them (ok, I do add RAM initially if I will be running gimp or such on it). The macs I usually buy low end machines on their 'end of model' sale and get a nice price on those as well.
Other than adding RAM or a new drive now and again I have only upgraded 1 machine in 20 years. I got a clocked doubled P180 for a P90 once. It wasn't really worth it. The rest of the machine was mostly obsolete and the machine just got bus bound. Paying a premium for upgradability is a waste of money in my book. Part of that comes from having multiple computers. I can always just demote all the machine, give away to oldest machine, and buy a new for the machine that needs CPU.
Re:Problems with induction - you barbarian!
on
Concept PC 2001
·
· Score: 2
My god! You still use floppies? I'm sorry, but you can't have a nice modern PC until you can get over that.:-)
Another example of the dark side of PCs. The inertia for `the way we have always done it' is immense.
When I got my first iMac I was worried about the no-floppy. It was pointless. I have never wanted a floppy for any mac since then. I still use them on PCs that are too old to boot from CD or net. Got to get linux on there somehow.:-)
Re:Only one problem...
on
Concept PC 2001
·
· Score: 5, Insightful
First off, design is cheap. Divide design cost by 1,000,000 computers and it vanishes. Second, problem is that PCs are designed. They are designed to minimize production costs while still being marketable. When people say paying for design they frequently mean paying for better usability, longevity, or aesthetics at the expense of low cost.
This can't be the product of a serious HP design effort...
Look at the pictures again and consider ergonomics.
In the first picture (tall skinny) your right knee is going to be bashing into the CPU.
In the second picture the helpful model has turned to face his cpu box which has no user interface funcations at all and will be twisting his head to the right all day to see the monitor.
3rd picture, not enough context to tell. Looks like it is back in knee bash position. Why is this cpu box taking up my desk? The CD is in the screen after all.
Now let's talk design. Just because this is different from the 20 year old PC form factor doesn't make it `designed'. Look at the display. Why is only 50% of the object's area useful display? Why is there a big handle on the bottom of it? I suspect it serves some other function, but it looks like a handle to me. Maybe I can hang my keyboard on the monitor handle? And no patententing the keyboard hanger HP, thats my idea.
I suspect we are not looking at a design effort, but rather some engineers were tasked to show what a bluetooth maximized PC would look like and produced a minimal vision.
Questions for future consideration...
Why isn't their a bluetooth headset there. Integrate my music grade headphones, my voice control mic, and my telephone. Give me voice dialing while you are at it.
Why isn't my PDA sitting there syncing through the bluetooth? Aren't we trying to sell a vision? show me!
Why am I looking at the same old cubical design? You just took away 16" of monitor depth yet you are showing me a cube designed with a corner desk to hold deep monitors. This monitor is incompatible with corner desks! Revisit that. Shrink the cube or use the space. $20/sqft/year. Use it or lose it.
I don't want to hassle with batteries. My freaking electric toothbrush charges inductively by sitting on a special base. Wacom powers their pointers and mice by wireless power transfer. Give me a little power mat I can place my keyboard an mouse on that will charge them. Make sure their battery can go a couple of days so I can forget once in a while, but let them recharge when idle. Consider solar power. It works for calculators in offices. Do my headset too while you are at it.
Where is my video conferencing camera? Stick it in my display like a little pointable eyeball. you've got enough room! and give me a little shutter I can flip over it so I don't feel watched all the time. maybe one of those round things in the display is a camera and one is a speaker?
Slashdot Mirror
Briefly...
It is a really well written desription, it is a shame Design Engineering didn't have an writer that could understand it.
The clothing and laundry machine industries are already gearing up to put rfid tags in garments. The idea is that your washer and dryer will watch what you put in them and then behave accordingly or warn you that you are about to turn all your underwear pink.
Oh look, I went searching for a link, and even found one on slashdot
When I see a link on the internet for massive bust I don't usually expect to get a page of text...
I'm in for $10. Maybe sending checks in care of his lawyer would be an appropriate mechanism?
Years ago the informatics crowd I hung with was working on the relationship between computation and power.
It turns out that because you need to represent state in a stable mechanism, and you can't change the stable state without using energy that there is a lower bound on the energy required to perform a given calculation.
I have forgetten everything else, and I suspect it is many orders of magnitude smaller than current cpu power usage, but it is enough to break the '==' operator on the parent's title.
Sapping heat from the smokestack contents will probably cause it to not work correctly.
The goal of a smokestack is to get the harmful exhaust away from the ground long enough that it disperses sufficiently before touching down.
This is done with convection. The hot gas in the tall stack creates the draw that powers it and blows the plume up after it leaves the stack, the hot plume continues to lift itself until it bleeds off too much heat, then it starts coming back down, but presumably dispersed enough to not be too noxious.
The smoke stack was designed with a known gas temperature and dispersal requirement and a desire to minimize masonry. If you take away heat from the gas you will reduce your plume altitude and cause it to come down in a more concentrated region.
I doubt you can use the thermo-generated electricty to run blowers to compensate. The `no free lunch' law of thermodynamics will probably forbid that. (Unless blowers are much more efficient than convection.)
Now, if you are just bleeding off waste steam then it would work, but most of the energy in steam is the expansion from water to steam, there is relatively little left in the puffy clouds.
Mostly unrelated note: I used to live in Pittsburg in a community where all the houses were required to have slate roofs, stone or brick exteriors and no wood trim. Even the window frames were metal. It was a fire-proof community from the days when the steel mills spewed lots of solids including hot cinders. The plume was powerful enough to carry those large distances fast enough that they were still hot enough to start a fire.
Just write yourself a little program to...
- wait a second
- start recording and start track X off the cd
- when the track ends stop recording
- trim the silence off the ends of the track
- encode the mp3
- repeat for all tracks
Ok, you will be ripping real time, but big deal. Let it go overnight. You will also need to type in your own track info until someone writes a new freedb-like service that uses a fuzzy audio signature instead of the digital signatures.No special hardware or loopback cables are required. (well, maybe one cable if your machine doesn't let you route CD audio to the DAC input) Just a different ripper than you are used to.
Wasn't clear to you? It might help to read the acual[sic] comment once in a while.
:-)
I'll be more clear. Basically no one has powered ethernet in their closets today, so you either have to buy the 3com power adding device mentioned in the article or replace all your ports with power supplying ports. For new build outs I presuming that powered port devices are more expensive than non-powered.
Ok, you want a hub. Go down to compusa and grab a 5 port 10/100 hub for $50. Save the $150 + install and if someone steals your hub buy another one.
I don't see much point in using a switch in your office, unless you really do a lot of high speed transfer between your cubicle machines or have an awful network. If you REALLY need a switch, then that will set you back $60 instead.
I wonder where Tom got the numbers for the cost comparison? He has cable pulling at $300+ for each run, I pay about $120. If his cable puller is charging him quad for four wires he needs a new cable puller. It ain't rocket science. He has upstream switch ports at $90ea! I'm paying $25 each. Then he has the mystical `gigabit switch port' row.
The maintanence/year row implies that the maintanence cost of the 3com device is zero. I would rather have something like "15% of cost"/year for it. I don't have many 4 year old hubs that haven't either died or got some blown ports. Lots of lightning around here. Its especially hard on hubs and NICs because of the ground surge differential on close strikes.
I see no indication of either the extra cost for powered ethernet devices or the electrical work to power the 3com devices.
The traditional wiring is costed for the worst case, where 4 ports are really active. Needing 4 ports available everywhere is not the same as needing 4 ports active everywhere. I routinely run four times the copper that I will need and activate ports as needed.
An even better comparison would have been to compare the 3com wall jack hub to a free standing hub. But then the 3com would not have looked so good.
It's a shame 3com didn't have these during the dotcom boom. They could have sold dumptruck loads of them. Now? I doubt it.
I have not made an ontology, but it seems to me that nearly all exploits the past few years have been (in decreasing prevalence order)
- data buffer overflow
- string overflow
- filename
.. abuse
A language with safe memory management will eliminate the first two. The third needs a more robust set of filename functions.Its not impossible, or even hard, to avoid these sins in C programming. But, it also isn't impossible or even hard to screw up and commit this sins.
Programmers make mistakes. That is why it is called programming instead of typing. Choosing a language that minimizes the security impact of mistakes makes a lot of sense.
Don't forget about other criteria. You may need the speed that can be had with well written C code. Usually you won't.
I look at my servers. They are all the slowest rackmount machines I could buy from Gateway when I bought them, 800MHz PIII is typical. (They are plural because the have different security policies, not because of load.) They handle things like mail, http, samba, cvs, ldap, the usual suspects for a 100 engineer software firm. They rarely go beyond 5% cpu utilization. I would gladly sacrifice my surplus cpu cycles for slower, safer, services. When they do go beyond 5% it is almost always for a very specific function like the rsync algorithms or blowing backup data over to another box. Make the hot spots of these functions fast, spend a lot of time making them secure. Probably not more than 400 lines of code between them. Let the rest be written in a safe language.
One other factor to consider is that the gzip transfer encoding compresses much better than the algorithm in the modem. Part of this is the algorithm with its larger dictionary size, the other part is the `pure' data stream being fed to it. It is just the html, not the html interspersed with ppp, ip, and tcp headers.
Speaking as the maintainer of a site that is periodically slashdotted...
e d, everyone waiting in queue gets a periodic update, at a certain point the load of generating the updates swamps the machine. I have to limit the number of people in queue.)
Yes, a throttling server is a great idea. If you recognize that there will always be a load too high for you to handle (10 requests per minute for my site, yes minute, it is a physical device), then you must either decide to deal with the load or let the load crush your machine.
Consider a typical web server. When it gets overloaded it slows down, each request takes longer to handle, there are more concurrent threads, overall efficiency drops, each request takes longer to handle.... welcome to the death spiral. (on my site-which-must-not-be-named-less-it-be-slashdott
The key decision is to determine how many concurrent threads you can handle without sacrificing efficiency and then reject enough traffic to stay under that limit.
This is where optimism comes in and bites you in the ass. You remember that every shunned connection is going to cost you money/fame/clicks whatever so you set the limit too high and melt down anyway.
It is 15 minutes after the article post and the site is dead. I got to the part about calculating how much RAM was required per visitor and multiplying by the expected number of visitors.
:-)
Maybe they need to adjust their constants.
It is those d*mn modem users that drive up the RAM use. They stay connected longer on their GET and tie up resources longer.
You are getting more info from your iPod than I am or the guy at neuron.com. Notice his kernel messages show an unknown type device and sequential ascii letters in all the text fields. This is the same thing I get. You on the other hand have it actually recognized as a SCSI disk!
I'll have to play around some more. I suspect my ohci1394 is goofed. Maybe I'll put debian on an imac so I can trace through that and see what is different.
Nice to know there is progress to be made!
I tried. I have firewire drives on my linux box. The iPod does not work as a disk `out of the box'. When the driver goes to read the `config' page it gets back garbage. Maybe there is a special command to flip the device into disk mode, maybe there is a bug in my linux 1394 stack, but it sure doesn't mount as a disk.
If I were smarter about OS X, I would figure out how to trace 1394 packets and just reverse engineer what happens on the Mac, then make that happen on Linux.
Fuel cells will win in efficiency. Probably by a large margin.
Perhaps turbines have other advantages...
20W won't get your a dual athlon. I measured my Celeron 700 machine at 70W and a 1.1GHz Athlon box at 110W. Maybe 130-150W for a dual Athlon. You will need a little bank of 8 turbines to power the beast.
And about flashlight life... cockeyed has a series on "how much is inside". Checkout the battery one for a shocker about how much is really in a pair of D cell batteries. Then try to figure out why your flashlight is always almost dead. Who comes in and uses up your flashlight?
"But where does the exhaust go?" - well, out, that is rather the definition of exhaust. The exhaust is water vapor, unused combustion air, and heat. That shouldn't be a problem. Well, you won't want 20W to 40W of heat running in your pocket, but other than that it should be fine.
"And isn't this kind of a step back in our attempt to stop using fossil fuels..." - I am an American. My goverment has no such policy. All your oil are belong to us. For those of you in conservation minded countries, hyrdrogen is not a fossil fuel. It is a theoretically handy way to store electrical energy. There are technical hurdles. Not insurmountable, just insurmounted. Once there is a demand, there will be a way of distributing and storing the hydrogen.
Which Apple partition destroying software would that be? I must have missed that one. I am only aware of two.
The iTunes partition destroyer was pulled in something like 24 hours and replaced not long after.
Some years ago there was a problem with certain models of hard drives (Quantums I believe) that didn't handle their write caches well on a scsi reset. That went on for quite a while, but was not an issue with supported Apple hardware, it was some 3rd party drives that had tweaks to enable write behind caching. (The very large Oracle installation on Alphas that I work with had the same problem with them. Unable to resolve it with the vendors we finally scrapped all the disks and replaced them with a different vendor's drives.)
Got one already! Not only cool, but its the only shirt in my closet that my girls won't steal. Its "too scary".
It seems we live in worlds too different to even communicate. Allow me to supply more detail...
:-)
Zip - it never entered my mind to use a zip disk. I have no use for cheap, unreliable storge (floppies), I certainly have no use for expensive unreliable storage.
Bootable CDs - my prefered way to install software. Fragility doesn't matter. I lose them or loan them. I just burn a new one each time I need one.
Bootable Network - I use this. If my NIC dies I replace it. I'm using Linux on these machines. I do not have a driver issue to deal with. If I did, I would just use a CD or CD-R to tote the driver over.
Everything is networked. All my valuable data moves around at night using rsync so it is stored on multiple machines. rsync is patient. Most users (image editors aside) don't make enough data that rsync won't catch up in a couple of hours at night, even on a modem.
You use "floppies" and "reliable" in the same sentence. I'll assume that is a typo.
You mention me using a costly horrible to upgrade box. I'm not sure what you mean there. I buy my wintel boxes for $400 (cheapest machine at BestBuy) and never upgrade them (ok, I do add RAM initially if I will be running gimp or such on it). The macs I usually buy low end machines on their 'end of model' sale and get a nice price on those as well.
Other than adding RAM or a new drive now and again I have only upgraded 1 machine in 20 years. I got a clocked doubled P180 for a P90 once. It wasn't really worth it. The rest of the machine was mostly obsolete and the machine just got bus bound. Paying a premium for upgradability is a waste of money in my book. Part of that comes from having multiple computers. I can always just demote all the machine, give away to oldest machine, and buy a new for the machine that needs CPU.
My god! You still use floppies? I'm sorry, but you can't have a nice modern PC until you can get over that. :-)
:-)
Another example of the dark side of PCs. The inertia for `the way we have always done it' is immense.
When I got my first iMac I was worried about the no-floppy. It was pointless. I have never wanted a floppy for any mac since then. I still use them on PCs that are too old to boot from CD or net. Got to get linux on there somehow.
This can't be the product of a serious HP design effort...
Look at the pictures again and consider ergonomics.
Now let's talk design. Just because this is different from the 20 year old PC form factor doesn't make it `designed'. Look at the display. Why is only 50% of the object's area useful display? Why is there a big handle on the bottom of it? I suspect it serves some other function, but it looks like a handle to me. Maybe I can hang my keyboard on the monitor handle? And no patententing the keyboard hanger HP, thats my idea.
I suspect we are not looking at a design effort, but rather some engineers were tasked to show what a bluetooth maximized PC would look like and produced a minimal vision.
Questions for future consideration...