I had to save the edited memory by breaking out to the monitor. (Giving the monitor the addresses of the screen buffer as the region to save.)
The bulk of the work was done by the monitor routines I borrowed.
No, this was not production, but it was simple, uncluttered, and shocked the socks off some engineers who saw it, until I showed them the actual code, and how much work it was to save and use the text. At which point we would have a good laugh.
The screen editor I wrote for my FORTH was actually useful, even though it was basically the same set of tricks with a FORTH underneath instead of a debug/monitor ROM.
Really, basic vi and emacs started out the same way, simple cursor functions on time of a programable base, lisp in the case of emacs, ed and then ex in the case of vi. Oh, and there was Unix under the programmable layer.
(Uhm, yes, at the time, 6800 machine code was, for me, a programmable layer. I sure spent a lot of time figuring out that the advantages of "higher" level languages was not in specific instructions or functions that were missing in the lower level so much as it was in the ability to borrow from the experience of other programmers through an agreed-on common language interface.)
Good editors, in my opinion, remain the same: a simple, well-integrated set of functions to move the cursor around a file, interfacing with a programming language that is able to edit the contents and put the results back in the visible edit buffer. Oh, and you need some glue functionality like the ability to select the target of an operation, but where that gets complicated is in keeping the visual feedback in sync with the actual operations.
He probably has, and has found that running the tools in the VM is more convenient for his work right now.
His point that when you choose the Mac you should be conscious that you are choosing to dig into the leading edges of the work of various dev teams is quite valid, however.
As is the implicit assertion that all the fun work is (precisely) out there on the edge.
I think you have just stumbled on the real reason Oracle is suing Google.
Oracle's incarnation of Sun wants Android for their very own and they aren't brave enough to try (again) to do it from scratch. Which may be why Google is having trouble being forthcoming with the source code for certain instances of Android.
If he's not "deploying to BSD" (his own words), the choice of the Mac was guaranteed to be just as problematic as the choice of the Lenovo, involving installing a foreign OS.
If he had taken the Lenovo, his first step was a wipe and an installation of the virtualization environment, then a re-installation of the MSWindows environment as a virtual environment. And his working Linux environment should also be virtual. Or, perhaps, he can install Debian on the metal and only install MSWindows virtual.
On the Apple, he has the additional option of using the OS the hardware was designed for as the base install and running Linux in a VM on the Mac OS. (And I would still run my working Mac OS as virtual, if I could talk purchasing into letting me get the RAM and disk.)
(Okay, I know some people who run Linux as a VM in MSWindows, but that's serious perversion.)
But we do deploy to (the Mach kernel on) BSD. Safari on MSWindows is yet another option that may need testing, but the percent of the general internet user population using the Mac now is not at all insignificant.
Fedora and Debian both provide a core repository and non-core repositories. I'm not familiar with Ubuntu.
OpenBSD does the same, and my memory is that freeBSD does, as well. The difference is primarily one of scale. openBSD, for instance intentionally keeps the core repository much smaller than most other distributions, and not just because the team is (intentionally) small.
The problem with the Mac, and it's a real problem, is that there is no official repository. Well, wasn't.
Now there is, but it's neither free nor open, as near as I can tell from the outside. (I'm not interested in being on the inside right now, so I'm judging it by the reports.)
But the app store still doesn't solve the underlying problem of dependency. Apple really let us down when they dropped the ball on this one. They have enough money, they could be supporting all three unofficial distros, so that you aren't as restricted by which package manager you've loaded a package by. And they could be encouraging, with financial encouragement, those guys to learn interoperability.
And they could set up a proper, funded, official commons coordinating organization to bring the three repositories and the app store together. With an official promise that they won't just drop it again the next time it's inconvenient to be having to put up walls between their project introduction hubris and the free world.
In the meantime, yeah, like lots of others are saying here, this Ted guy seems to be kind of young for shifting to old codger mode.
There is a tendency to see, in re-programmability, the solution to all the worlds problems.
There is a tendency to see, from that, the mirage of a "one true path", and a tendency to identify the current favorite tool as "the one true tool".
After he gains a bit more experience, he will discover that the one true path for mortals is context dependency. Then he won't be so surprised when tool X has limits, and he will be able to see the options better. No big deal.
Mac is another option. It's important to learn and use a variety of tools, and not lock yourself into one. Using Apple's stuff is just as important as using Microsoft's stuff. (... as using Red Hat's, as using Google's,...)
Which is why the whole which-editor-is-best argument is a lot of noise, useful noise in some cases, not so useful in others.
(Even though I personally don't seem to be able to get comfortable with Emacs. Maybe I should learn lisp.)
Maybe he has got ahold of one of the new(er) GUI emacs bundles.
Like XEmacs or GNU Emacs.
They just feel unnatural to me, probably because I started with TRS80 Basic, bounced from there to a screen editor I wrote myself in a few lines of 6800 assembler borrowing functions in the TVBug (I think it was) monitor on the microchroma 68 prototyping board, to vi, to WordPerfect, to "brief", and then another screen editor I wrote in a variant of FORTH I wrote in 6809 assembly language (TRS80 Color Computer).
vi was probably the most stable text editor I used, from '83 until I discovered BBEdit and the Metrowerks IDE editor on the Mac in the mid-90s.
(I currently seem to be mostly using Apple's Project Builder built-in editor on the Mac, GEdit on Linux, and Hidemaru on MSWindows.)
Whenever I use emacs, I still find myself longing for that old FORTH screen editor. (That may have something to do with why I liked WordPerfect's text editor, although WordPerfect, at the time, was actually quite useable for editing text, if you remembered to save as text.)
The rant on TextMate that the guy links to mentions BBEdit as passable, but not oriented enough to Ruby, then praises NetBeans.
I myself like NetBeans, for Java. Not perfect, but there is no perfect editor.
(Except the one in my head, the one I will definitely write once I get my perfect re-write of FORTH finished.)
What she might really be asking for is some of your time.
First, make some work dates with her, where you are both working in the same room for some extended time. When she has problems, don't tell, ask. Get her opinion, write it down in a notebook so you can refer to it later when you go shopping, then say, "... well, anyway, here's how I do it." Be careful not to say or imply she has to do it that way.
(This can be hard with people close to you, you may have to re-think how you are interacting with her. But it's worth the effort.)
Now, if she really has issues with the Mac UI, the next step is to max the RAM on it, get a virtualization environment, and install MSWindows on it. Preferably in a VM, if the machine supports virtualization.
But you're still not done. You'll still need to be around to help her over the speedbumps.
Well, if it's clear that the peculiarities of the Mac keyboard are an issue, you may need to skip that step.
The 3rd step is to make getting the new machine for her a series of dates, both on the web and in real stores. Let her choose what she looks at and tries out, make sure she gets hands-on time with the software she uses most, and do your homework after you get home with literature and the notes you took. (Don't forget the notebook. At least think about paper and pencil instead of the MacBook.) You'll know what to check on by what she shows interest in.
4th step -- don't just drop the dates after she has her new machine. The new versions of the software, all of that is going to be more speedbumps. Besides,regularly giving your wife some of your time is a good habit to keep. Rewarding for you, too.
Make buying the new notebook an excuse for shopping dates. Let her make as many of the decisions as she will, except, when she settles on something she really likes, max the RAM and HD like everyone says.
Then take it home, install Linux on it and re-install MSWindows virtualized in that. Erm, well, since the guy who is asking doesn't mention using Linux himself, maybe not that.
Anyway, after they get it home, he needs to plan time he can be around when she's working, so he can help her over the speedbumps.
(Actually, I would start and end with making sure he's around to help her. If he forgot that part when he gave her his old Mac, that's part of the problem.)
Actually,what she probably wants is a little face time with him. I'm not talking about how close, just that he probably mostly needs to give her a little more of his time.
So, first, he should be their while she's working on the thing, and answer questions and ask questions.
Second, shoot a little cash on an OEM copy of MSWindows, max out the RAM, install MSWindows virtualized, and let her use that while they do the third step.
Third step, if it's necessary, make some dates to go shopping with her, on-line and in brick-and-mortar stores. Get her some brochures. Hold her hand while she's trying things in the stores.
One question I'd ask before even suggesting installing MSWindows on the older Mac:
Did he just give her the Mac and let her try to figure it out for herself, or is he making sure to be in calling distance part of her working time? Even if installing MSWindows virtualized on the old Mac (if it supports it), he should be sure that he is available to get her over at least some of the speed bumps she hits. Maybe make a work date where they work together.
A little TLC goes a long ways to help people handle context and paradigm shifts.
Helping a person choose for him or herself is the best general approach I've found when people ask me for advice on buying computers. (Well, when I feel obliged for some reason to help.)
But choosing from pictures is only good for the wife that likes catalog shopping. Hands-on is always a good thing.
Nope. Still can't believe I'm reading this. You really do have yourself fooled.
Is it worth replying to this?
You tell me, if everyone in the world "opened their eyes" as you suggest, who is going to move packages through the parcel post system for one third of what you think is minimum wage? Or are you really saying that we should treble the prices of parcel post so that everyone can meet your minimum wage when they realize the gap between your world and theirs? And don't you tell me those guys are lazy. If you do, I'll curse you with having to work part time sorting parcels in the middle of the night like I have to pay rent.
Teachers? Are you going to fork over the taxes or whatever to pay elementary teachers your minimum wage when they realize just how much finer you are than them, because you and all your friends are "working hard" scraping a few points each of you out of every single dollar they get paid, so that they can pay ever more taxes to pay for the "mistakes" you guys make? And remember, these are the guys trying to keep the rising generation motivated to learn enough skills to do something besides watch TV all day, when you talk about being lazy.
Not every low-paying job is slinging hamburgers over at MacDonalds, but even that is real work.
Look at yourself in the mirror when you talk about not understanding the fundamental idea of unintended consequences.
Look at yourself in the mirror when you talk about rules and how they don't really match what people do.
Look at yourself in the mirror when you talk about why more and more new rules end up having to be made, and then made more complicated.
Some of what is done in the financial sector is valuable to society in general. But nowhere near matching the value you guys are sucking out of society to feed your illusions of grandeur.
Get a job. Get a real job, and work it for a few years so you can understand what real work is. Then maybe we can talk about your theories of how to make everyone wealthy.
Re-calculate the future value of your current wage in a society in which you have sucked all the value out and left your "rich" (spoiled, counter-motivated) kids with a tax burden beyond the ability of 400 million extreme over-achievers to pay back.
(And ask yourself who your kids are going to owe that tax burden to.)
Seriously, you are not just part of the problem, you are one of the enablers of the problem.
And if you think poor people are just lazy, I have a pyramid scheme or a dozen that I could sell you right away, I'm sure.
It also happens to be a conglomeration of languages.
I must admit, I was wondering, when I read the summary, whether he meant one specific.net language, the underlying run time which shines (murkily) through all the.net languages, or the dev environment (No, don't waste your breath on that quibble, the environment is, mathematically, a language.), the implicit glue language(s) which hold the framework together, the framework, or what. I read the friendly article, and it's clear he means all the above.
I wouldn't want to work with him, I'm guessing, because my family means something to me. I'm passionate about coding, But I can hold concentration without sacrificing my sleep and my family, and I don't do arbitrarily short deadlines just to pull some VC backer's back side out of the fire.
VC is a matter of luck, not some innate virtue. I don't like to see people suffer, but market windows are, in the end, almost always self-induced mirages. The deserve what they get when they do that.
Guys who put arbitrary deadlines on schedules are precisely the sort of people who got companies like Microsoft started, and if they succeed, they tend to go down the same path toward mediocrity.
Meaningless competition is precisely the reason the world is heating up.
That said, these days, I spend more of my time coding in natural languages, and my target systems are a bit softer and wetter than the systems that run the languages either you or this David Barrett guy are talking about.
I just lost a USB flash drive. Tried everything I could to recover it, and it's just dead.
It did warn me before it died, Japanese characters in file names turning into question marks and such. I think I got all my data off of it and into a tarball, but I'm not sure.
I think the device was still under warranty, but it's going to cost me enough to send it in and get it fixed (by replacement, I'd guess), and Buffalo is not going to recover my data for me, so why bother?
I'm always having problems with USB. Not to mention that the connector slips out awfully easily. I used to think it made an OK replacement for the floppy disk drive, but not now.
USB does not strike me as something you want to have important data riding on.
Slashdot Mirror
I had to save the edited memory by breaking out to the monitor. (Giving the monitor the addresses of the screen buffer as the region to save.)
The bulk of the work was done by the monitor routines I borrowed.
No, this was not production, but it was simple, uncluttered, and shocked the socks off some engineers who saw it, until I showed them the actual code, and how much work it was to save and use the text. At which point we would have a good laugh.
The screen editor I wrote for my FORTH was actually useful, even though it was basically the same set of tricks with a FORTH underneath instead of a debug/monitor ROM.
Really, basic vi and emacs started out the same way, simple cursor functions on time of a programable base, lisp in the case of emacs, ed and then ex in the case of vi. Oh, and there was Unix under the programmable layer.
(Uhm, yes, at the time, 6800 machine code was, for me, a programmable layer. I sure spent a lot of time figuring out that the advantages of "higher" level languages was not in specific instructions or functions that were missing in the lower level so much as it was in the ability to borrow from the experience of other programmers through an agreed-on common language interface.)
Good editors, in my opinion, remain the same: a simple, well-integrated set of functions to move the cursor around a file, interfacing with a programming language that is able to edit the contents and put the results back in the visible edit buffer. Oh, and you need some glue functionality like the ability to select the target of an operation, but where that gets complicated is in keeping the visual feedback in sync with the actual operations.
Unicode does make things a little more difficult.
He probably has, and has found that running the tools in the VM is more convenient for his work right now.
His point that when you choose the Mac you should be conscious that you are choosing to dig into the leading edges of the work of various dev teams is quite valid, however.
As is the implicit assertion that all the fun work is (precisely) out there on the edge.
I think you have just stumbled on the real reason Oracle is suing Google.
Oracle's incarnation of Sun wants Android for their very own and they aren't brave enough to try (again) to do it from scratch. Which may be why Google is having trouble being forthcoming with the source code for certain instances of Android.
(The obligatory)
You must be new here!
(since no one else seems to have said it yet.)
If he's not "deploying to BSD" (his own words), the choice of the Mac was guaranteed to be just as problematic as the choice of the Lenovo, involving installing a foreign OS.
If he had taken the Lenovo, his first step was a wipe and an installation of the virtualization environment, then a re-installation of the MSWindows environment as a virtual environment. And his working Linux environment should also be virtual. Or, perhaps, he can install Debian on the metal and only install MSWindows virtual.
On the Apple, he has the additional option of using the OS the hardware was designed for as the base install and running Linux in a VM on the Mac OS. (And I would still run my working Mac OS as virtual, if I could talk purchasing into letting me get the RAM and disk.)
(Okay, I know some people who run Linux as a VM in MSWindows, but that's serious perversion.)
But we do deploy to (the Mach kernel on) BSD. Safari on MSWindows is yet another option that may need testing, but the percent of the general internet user population using the Mac now is not at all insignificant.
Fedora and Debian both provide a core repository and non-core repositories. I'm not familiar with Ubuntu.
OpenBSD does the same, and my memory is that freeBSD does, as well. The difference is primarily one of scale. openBSD, for instance intentionally keeps the core repository much smaller than most other distributions, and not just because the team is (intentionally) small.
The problem with the Mac, and it's a real problem, is that there is no official repository. Well, wasn't.
Now there is, but it's neither free nor open, as near as I can tell from the outside. (I'm not interested in being on the inside right now, so I'm judging it by the reports.)
But the app store still doesn't solve the underlying problem of dependency. Apple really let us down when they dropped the ball on this one. They have enough money, they could be supporting all three unofficial distros, so that you aren't as restricted by which package manager you've loaded a package by. And they could be encouraging, with financial encouragement, those guys to learn interoperability.
And they could set up a proper, funded, official commons coordinating organization to bring the three repositories and the app store together. With an official promise that they won't just drop it again the next time it's inconvenient to be having to put up walls between their project introduction hubris and the free world.
In the meantime, yeah, like lots of others are saying here, this Ted guy seems to be kind of young for shifting to old codger mode.
I never use the stock OS install anyway, so I just click the case sensitive box when I format the HFS+ partitions.
There are other options.
But then you have to do some really hard work to get the Mac environment to test on.
There is a tendency to see, in re-programmability, the solution to all the worlds problems.
There is a tendency to see, from that, the mirage of a "one true path", and a tendency to identify the current favorite tool as "the one true tool".
After he gains a bit more experience, he will discover that the one true path for mortals is context dependency. Then he won't be so surprised when tool X has limits, and he will be able to see the options better. No big deal.
Mac is another option. It's important to learn and use a variety of tools, and not lock yourself into one. Using Apple's stuff is just as important as using Microsoft's stuff. (... as using Red Hat's, as using Google's, ...)
Which is why the whole which-editor-is-best argument is a lot of noise, useful noise in some cases, not so useful in others.
(Even though I personally don't seem to be able to get comfortable with Emacs. Maybe I should learn lisp.)
Maybe he has got ahold of one of the new(er) GUI emacs bundles.
Like XEmacs or GNU Emacs.
They just feel unnatural to me, probably because I started with TRS80 Basic, bounced from there to a screen editor I wrote myself in a few lines of 6800 assembler borrowing functions in the TVBug (I think it was) monitor on the microchroma 68 prototyping board, to vi, to WordPerfect, to "brief", and then another screen editor I wrote in a variant of FORTH I wrote in 6809 assembly language (TRS80 Color Computer).
vi was probably the most stable text editor I used, from '83 until I discovered BBEdit and the Metrowerks IDE editor on the Mac in the mid-90s.
(I currently seem to be mostly using Apple's Project Builder built-in editor on the Mac, GEdit on Linux, and Hidemaru on MSWindows.)
Whenever I use emacs, I still find myself longing for that old FORTH screen editor. (That may have something to do with why I liked WordPerfect's text editor, although WordPerfect, at the time, was actually quite useable for editing text, if you remembered to save as text.)
The rant on TextMate that the guy links to mentions BBEdit as passable, but not oriented enough to Ruby, then praises NetBeans.
I myself like NetBeans, for Java. Not perfect, but there is no perfect editor.
(Except the one in my head, the one I will definitely write once I get my perfect re-write of FORTH finished.)
A little bit late to be posting this, but, ...
What she might really be asking for is some of your time.
First, make some work dates with her, where you are both working in the same room for some extended time. When she has problems, don't tell, ask. Get her opinion, write it down in a notebook so you can refer to it later when you go shopping, then say, "... well, anyway, here's how I do it." Be careful not to say or imply she has to do it that way.
(This can be hard with people close to you, you may have to re-think how you are interacting with her. But it's worth the effort.)
Now, if she really has issues with the Mac UI, the next step is to max the RAM on it, get a virtualization environment, and install MSWindows on it. Preferably in a VM, if the machine supports virtualization.
But you're still not done. You'll still need to be around to help her over the speedbumps.
Well, if it's clear that the peculiarities of the Mac keyboard are an issue, you may need to skip that step.
The 3rd step is to make getting the new machine for her a series of dates, both on the web and in real stores. Let her choose what she looks at and tries out, make sure she gets hands-on time with the software she uses most, and do your homework after you get home with literature and the notes you took. (Don't forget the notebook. At least think about paper and pencil instead of the MacBook.) You'll know what to check on by what she shows interest in.
4th step -- don't just drop the dates after she has her new machine. The new versions of the software, all of that is going to be more speedbumps. Besides,regularly giving your wife some of your time is a good habit to keep. Rewarding for you, too.
Make buying the new notebook an excuse for shopping dates. Let her make as many of the decisions as she will, except, when she settles on something she really likes, max the RAM and HD like everyone says.
Then take it home, install Linux on it and re-install MSWindows virtualized in that. Erm, well, since the guy who is asking doesn't mention using Linux himself, maybe not that.
Anyway, after they get it home, he needs to plan time he can be around when she's working, so he can help her over the speedbumps.
(Actually, I would start and end with making sure he's around to help her. If he forgot that part when he gave her his old Mac, that's part of the problem.)
Actually,what she probably wants is a little face time with him. I'm not talking about how close, just that he probably mostly needs to give her a little more of his time.
So, first, he should be their while she's working on the thing, and answer questions and ask questions.
Second, shoot a little cash on an OEM copy of MSWindows, max out the RAM, install MSWindows virtualized, and let her use that while they do the third step.
Third step, if it's necessary, make some dates to go shopping with her, on-line and in brick-and-mortar stores. Get her some brochures. Hold her hand while she's trying things in the stores.
Let her make the final decision.
One question I'd ask before even suggesting installing MSWindows on the older Mac:
Did he just give her the Mac and let her try to figure it out for herself, or is he making sure to be in calling distance part of her working time? Even if installing MSWindows virtualized on the old Mac (if it supports it), he should be sure that he is available to get her over at least some of the speed bumps she hits. Maybe make a work date where they work together.
A little TLC goes a long ways to help people handle context and paradigm shifts.
Helping a person choose for him or herself is the best general approach I've found when people ask me for advice on buying computers. (Well, when I feel obliged for some reason to help.)
But choosing from pictures is only good for the wife that likes catalog shopping. Hands-on is always a good thing.
The GP's advice (pick by picture) is not so good, in general.
But it may work well with the GP and his wife (or girlfriend, I suppose).
Women are all different.
because she wants him too.
Get it?
Women aren't rational creatures. Period.
Pardon me while I pick my jaw up off the floor.
Nope. Still can't believe I'm reading this. You really do have yourself fooled.
Is it worth replying to this?
You tell me, if everyone in the world "opened their eyes" as you suggest, who is going to move packages through the parcel post system for one third of what you think is minimum wage? Or are you really saying that we should treble the prices of parcel post so that everyone can meet your minimum wage when they realize the gap between your world and theirs? And don't you tell me those guys are lazy. If you do, I'll curse you with having to work part time sorting parcels in the middle of the night like I have to pay rent.
Teachers? Are you going to fork over the taxes or whatever to pay elementary teachers your minimum wage when they realize just how much finer you are than them, because you and all your friends are "working hard" scraping a few points each of you out of every single dollar they get paid, so that they can pay ever more taxes to pay for the "mistakes" you guys make? And remember, these are the guys trying to keep the rising generation motivated to learn enough skills to do something besides watch TV all day, when you talk about being lazy.
Not every low-paying job is slinging hamburgers over at MacDonalds, but even that is real work.
Look at yourself in the mirror when you talk about not understanding the fundamental idea of unintended consequences.
Look at yourself in the mirror when you talk about rules and how they don't really match what people do.
Look at yourself in the mirror when you talk about why more and more new rules end up having to be made, and then made more complicated.
Some of what is done in the financial sector is valuable to society in general. But nowhere near matching the value you guys are sucking out of society to feed your illusions of grandeur.
Get a job. Get a real job, and work it for a few years so you can understand what real work is. Then maybe we can talk about your theories of how to make everyone wealthy.
Re-calculate the future value of your current wage in a society in which you have sucked all the value out and left your "rich" (spoiled, counter-motivated) kids with a tax burden beyond the ability of 400 million extreme over-achievers to pay back.
(And ask yourself who your kids are going to owe that tax burden to.)
Seriously, you are not just part of the problem, you are one of the enablers of the problem.
And if you think poor people are just lazy, I have a pyramid scheme or a dozen that I could sell you right away, I'm sure.
Think again, man. Think again.
It also happens to be a conglomeration of languages.
I must admit, I was wondering, when I read the summary, whether he meant one specific .net language, the underlying run time which shines (murkily) through all the .net languages, or the dev environment (No, don't waste your breath on that quibble, the environment is, mathematically, a language.), the implicit glue language(s) which hold the framework together, the framework, or what. I read the friendly article, and it's clear he means all the above.
I wouldn't want to work with him, I'm guessing, because my family means something to me. I'm passionate about coding, But I can hold concentration without sacrificing my sleep and my family, and I don't do arbitrarily short deadlines just to pull some VC backer's back side out of the fire.
VC is a matter of luck, not some innate virtue. I don't like to see people suffer, but market windows are, in the end, almost always self-induced mirages. The deserve what they get when they do that.
Guys who put arbitrary deadlines on schedules are precisely the sort of people who got companies like Microsoft started, and if they succeed, they tend to go down the same path toward mediocrity.
Meaningless competition is precisely the reason the world is heating up.
That said, these days, I spend more of my time coding in natural languages, and my target systems are a bit softer and wetter than the systems that run the languages either you or this David Barrett guy are talking about.
I just lost a USB flash drive. Tried everything I could to recover it, and it's just dead.
It did warn me before it died, Japanese characters in file names turning into question marks and such. I think I got all my data off of it and into a tarball, but I'm not sure.
I think the device was still under warranty, but it's going to cost me enough to send it in and get it fixed (by replacement, I'd guess), and Buffalo is not going to recover my data for me, so why bother?
I'm always having problems with USB. Not to mention that the connector slips out awfully easily. I used to think it made an OK replacement for the floppy disk drive, but not now.
USB does not strike me as something you want to have important data riding on.
Hmm. Of course, I don't suppose there are a lot of SSDs hanging off of serial scsi ports.
Maybe?
You want Intel to own your pipes, don't you?