You are right; the WinForms stuff, specifically are not only protected (AFAIK), they aren't even functional outside a Win32 environment (they more or less just wrap the Win32 api). Of course, that means a lot less than you may think. The portability is likely to be important mostly for server stuff, where you won't have an UI anyway. And as we've already been seeing, apps written under mono will tend to use GTK anyway.
I don't really think the promise of portability bewtween Win and Linux is the important part of mono. It is rather that the system is a pretty clean, well-designed one. Also, it _does_ offer excellent portability between Linux versions - run the same binary on whatever distro, on whatever hardware. Redhat on x86, Linux on an iPaq, Debian on a Sparcstation, RH on an IBM s390 - it will just work, without recompiling or installation issues.
The core of.net is a standard, and the IP is offered royalty free. MS is unlikely to be able to change that. Beyond the core, it would be nice to keep compatibility wherever it makes sense to do so, but if MS makes a fuss, just dump the pieces they want to keep to themselves. Mono doesn't live or die on 100% compatibility the way Wine does, for example.
In fact, given Linux' steadily increased prescence as a server, if MS goes off and makes mono incompatible with their own version (whether by API changes, implementation secrets or licensing stupidity), chances are developers who use.net for server stuff will decide to use the mono equivalents instead (since they are feee to move over to the windows side). WIth enough mono deployments, MS may well find itself locked in from raising too much of a fuss. But again, the real benefit of mono doesn't lie there anyway.
Well, yes and no. Clocks, battery meters, and other status stuff certainly counts as separate apps, they are being piled on onto PalmOS and such systems as well, just not as cleanly and safely.
Other examples include popping up an on-screen keyboard; or blanking the screen with a clock display when not in use. You still have the issues with multiple windows (in the X sense). Oh, and with a real OS you may well want to have several apps running and switch between them, even though you are only looking at one at a time.
For some of the newer devices, like the Zaurus, the screen is starting to become big enough that it also starts to make sense to actually have windowed applications. The Zaurus has the same resolution as my old laptop, and windowed apps worked fine there.
Oh, while we're on it - I've looked at screenshots from several projects. They generally do the right thing in using the stuff full screen. But then they pile on toolbars, status bars and big honking scrollbars and frames to the point where the actual application area is only half of the total screen estete. Ugh. That is the one thing I find Palm does very well - there is very little wasted screen estate on a palm device.
You know, people have all sorts of reasons for using a given OS other than ideological ones. For example, I'd like a Zaurus with Linux and a GTK-based interface. I would not choose a Zaurus over Palm or iPaq for ideology, but simply because I like the Zaurus form factor a lot better. I would like Linux over PalmOS or WinCE not because of ideology, but because I find it a comfortable and powerful system that I know well. I would like GTK over Qt (ora any of the dozen other toolkits), not because of ideology, but because I again like the GTK programming model, know it fairly well, and have some stuff written with GTK I would enjoy having on a PDA.
Had PalmOS or WinCE offered me a better platform for my needs, I would want them instead.
As for "out of work software engineers" - to be blunt, do I care? Nope. Turn it around: if Palm or anyone else chooses Linux over another system, they do so because it benefits their bottom line. In other words, staying with the previous solution gives the customer a more expensive or less good system. As a customer, I don't really feel the need to sponsor a commercial development team on another continent to the detriment of myself.
To put it even blunter: If they (as a team, not individuals) can't produce value over time at least equal to what they cost the company, they aren't doing the right job in the first place. The company has them doing the wrong job, or the kind of work they are doing is not viable for a commercial entity in the first place.
Something like Tiny X has a total footprint of 700k or thereabouts. X isn't (or doesn't have to be) that resource intensive. Hopefully everybody realizes by now that the "memory hungry" myth is just that; a result of unavoidably disingenoius reporting by system tools. And you don't _need_ a large dynamic font system with AA, or all the modules you load at run-time either, if you want to shrink resource usage further.
And X-less methods rack up that kind of resource use anyway; _something_ has to handle expose and redraw stuff, for instance, if you are aiming for something able to run more than one thing at a time. Apparently (I have not worked on it myself), qtopia requires the applications to handle WM-stuff by themselves.
I agree that Palm did a pretty good thing with their system, by only allowing one - full-screen - application to run at any one time. Easy model to handle, and resource efficient. But as the huge collection of hacks show, it is also quite limiting.
We have PDA:s that can also make cellular phone calls. We have phones that can double up as PDA:s. They seem to aim for exactly the same market, but, of course, they don't, since they're best features are aimed at different uses.
Same thing with calculators. I'd love to have a HP calculator that will also function reasonably as a PDA. I'm a lot less interested in a PDA that can also do some calculator functions.
It's all about where the focus is. Take the keyboard as an example: a dinky on-screen keyboard, or aphanumeric keyboard just isn't nearly as functional and convenient as a 'real' calculator keyboard a'la my deeply missed HP15, where all the functionality is right there, at your fingertips. Likewise, a phonepad isn't really that good for PDA functionality, and a touch screen isn't really that good for a phone.
Also, the software for PDA:s are of varying, and unknown, quality. One thing that really made the HP line of calculators stand out was their attention to various corner cases. When you got a result, you knew that was the correct one, to the practical limit of the hardware and encoding used. The Palm calculators I've tried have inevitably had various bugs and have missed special cases that made you get the wrong result from time to time - they would not handle over/underflow correctly in all cases, or use algorithms that would not give the stated precision over all of it's range, and so on.
My dream would be a new HP calculator with the format and design of the HP15c, but modernized (faster CPU with more memory; pisel screen, rather than segment, and so on). That one was a nearly perfect unit for me. After fifteen years, I had unfortunately dropped it, spilled coffee and soda in it, buried it under piles of books, stuffed it in dirty, dusty bags and submerged it too many times and it gave up:(
Even assuming it _is_ just a recompile that's needed (and, especially for OS:es that is far from the case), you still need to convince those holding the source to actually do it. If MS doesn't see a profitable enough market, they will not "recompile for a different platform", and you are sans Windows for your platform. If Oracle doesn't want to bet on the new architecture, you won't have your database available. The same goes for much of the rest of mainstream computing today.
OSS is interesting, as it - like Unix before it - partly changes that equation. With the source out there, a chip maker can do the port themselves, without having to rely on another company to be good to them.
Intel _has_ tried to introduce different platforms in the past - they had a RISC chip going for some years, for example, that never went anywhere. But, no platform ==> no third-party software ==> no system builders to pick it up.
Oh, and both AMD:s and Intel's 64 bit offerings _do_ already provide exactly that backwards bridging you're talking about. So where's the problem?
I went to school in Sweden, yes; I got my master's in CS. I thought briefly about maybe taking a bachelor's in mathematics, but alas, I like math more than math likes me:) I'm just not really focused enough to be able to do good work in the subject.
Multiple choice exams are bad wherever they are used; in a subject like mathematics, it's just ridiculous. In our introductory semester of mathematics, we were 150 students (it was mandatory for those planning to major in CS, physics, mathematics or statistics), and they still managed to have both a written exam and an oral exam for everybody that passed the written one. Granted, there were not nearly 150 people passing the written one, but anyway.
Most math departments have a course somewhere after the introductory sequence which teaches basic proof techniques often by studying the definition of numerical systems from logical axioms.
_After_ the "introductory sequence"? We started right off, in both introductory algebra and calculus I, with the basic proofs in each subject, and the emphasis was all along on understanding the theorems, to the point that the exams are in two parts: one ordinary test, and, for those that pass, a two hour oral exam about the theory.
Not the easiest way to get into the subject, perhaps, but you really do get a very solid theoretical foundation to pin your understanding on.
That said, I don't like calculus... The proof techniques are far too often of the type "half a page of dreary symbol manipulation, apply the magic trick that you would never think of yourself, and another half a page of manipulation". Discrete mathematics (and topology) tends to be a lot, well, cleaner and more beautiful.
There are a lot of issues one can bring up with the test - not identical versions of various software; different X drivers, one distro will have patches missing in the others and so on. Clearly, that greatly influences the results.
And that is a good point to take home. Optimizing compiles is _not_ the panacea for speed and responsiveness that - a minority, I believe - of source-based distros tend to bring up. There are so many other factors intimately involved in it that any benefits are generally lost in the noise.
For some specific components, it can be a good idea - but for those, most distros tend to ship several optimized versions that the installer chooses between at installation time.
Another domain that benefits are specialized, compute-intensive applications; things like simulators or other technical stuff. But then, those apps are generally tweaked and compiled by their users no matter what distro is used anyway.
I don't think that is the kind of problem the previous poster had in mind. Take the example with having a row of tiny buttons with less-than-clear icons on them.
Their functionality - or lack of it - has nothing to do with the problem domain. No matter what the application, they are making it harder to use than would otherwise be the case. They are difficult to see, to distinguish between, and easier to make an accidental "misclick" and select the wrong function.
I fully agree that an UI should be designed for the target user population. A desktop (whether Gnome, KDE, Windows, Mac or whatever) tends to be rather plain an uninnovative, and it must be as the target population is, well, everybody. A smaller, specialized segment, such as Blender users, can motivate a much more specialized, targeted UI as well.
But there are also some things that make using an app harder than it should be, no matter what the target population is (assuming it's at least human, and alive). They are things that stretch the sensory or motor-capabilities of the user too much (like finding the right button, when they are too small and too many, bunched together too tight), or stuff that could be one way or another without loss in functionality, and the "backwards" way is chosen. Those kinds of things would be a pure positive to change, with no downside for anyone.
Um, so? I sat on a P2/233 up until this year in the department. With 256Mb memory, running Gnome2 with Mozilla, AbiWord and Evolution was no problem at all. For office/productivity apps, the CPU simply isn't a serious bottleneck; the critical parameter is memory (to avoid swapping).
And why did I sit on such an old machine? Because I was too lazy to get up and order a new one until after New years, since it basically worked fine for what I used it for. Only when it started to act up (hard drive read failure rate was shooting up alarmingly, and the video card was going south) did I feel the need to get a new machine.
It's not about wear and tear. The feel of the thin, flimsy cotton or linen in dress shirts literally makes my skin crawl. If find the feel vaguely repugnant. And no, a thick weave (and I mean a weave with thick strands) of the kind I like is fairly loose, so it lets through as much air (if not more) than the thin - but tight - weave of a dress shirt.
And I have no problem finding shirts of the kind I like that fit otherwise. Sleeve length, width, shoulder play; all doable, except for the collar. The basic problem really is that while dress shirts are sold with various collar and arm sizes, these kind of shirts tend to be one-dimensionally sized - you get bigger collar, you get bigger everything else.
Fortunately, I work in academia, where jeans and t-shirt is perfectly acceptable dress day-to-day dress. Have a shirt, and you're well dressed. Throw on a jacket, and people wonder if I'm due at a wedding or something:)
I have a problem with shirts. I have a pretty thick neck, so most shirts I look at won't fit. If the neck is wide enough, the arms are far too long. The only recourse would be dress shirts, but, frankly, I hate them. They are only white (=sweat stains galore after a whole day) or baby blue, and are made in, in my taste, way too flimsy material.
I like shirts that use a thick, comfortable, material - heavy unsmoothed linen is fine, for example, or a _really_ heavy cotton weave. I also prefer darker, muted colors. But I seem unable to find such shirts with the neck width I need. Thus I use them, but unbuttoned, and without a tie.
They're just about level with Gamecube, actually, and both are very distant thirds to PS/2. The only market where X-box is going anywhere is the US; sales are not good in Europe, and pretty much disastrous in Asia.
Nah, they're not a monolith. If they were, they'd be in orbit around Jupiter right now instead of producing software. The real test of course comes in 2010 when we find out if MS will eat Jupiter or not.
In some respects.net is more open than Java. This has mostly to do with the core being accpted as a standard, which Sun repeatedly has refused to do with Java.
Lately there have been noises that Sun considers _any_ implementation of Java as subject to their licensing and control. This they can do since they own IP and patents on the Java architecture, and since they haven't submitted it as a standard (which would require opening up the licensing). This could not happen with the core.net architecture (and observe that I am talking about the standardized core, not the entire class library).
Re:Making life easy for a hacker
on
Nat Demos Dashboard
·
· Score: 3, Interesting
The apps choose what to send cluepackets about. I doubt any app writer would be dumb enough to send out an entered password as a cluepacket, anymore than they would print it in clear text on the screen.
Maybe explaining the reason for it to them? Most people can readily identify with this issue, and framing it as a way to get around it will likely be fairly well received.
Slashdot Mirror
You are right; the WinForms stuff, specifically are not only protected (AFAIK), they aren't even functional outside a Win32 environment (they more or less just wrap the Win32 api). Of course, that means a lot less than you may think. The portability is likely to be important mostly for server stuff, where you won't have an UI anyway. And as we've already been seeing, apps written under mono will tend to use GTK anyway.
.net is a standard, and the IP is offered royalty free. MS is unlikely to be able to change that. Beyond the core, it would be nice to keep compatibility wherever it makes sense to do so, but if MS makes a fuss, just dump the pieces they want to keep to themselves. Mono doesn't live or die on 100% compatibility the way Wine does, for example.
.net for server stuff will decide to use the mono equivalents instead (since they are feee to move over to the windows side). WIth enough mono deployments, MS may well find itself locked in from raising too much of a fuss. But again, the real benefit of mono doesn't lie there anyway.
I don't really think the promise of portability bewtween Win and Linux is the important part of mono. It is rather that the system is a pretty clean, well-designed one. Also, it _does_ offer excellent portability between Linux versions - run the same binary on whatever distro, on whatever hardware. Redhat on x86, Linux on an iPaq, Debian on a Sparcstation, RH on an IBM s390 - it will just work, without recompiling or installation issues.
The core of
In fact, given Linux' steadily increased prescence as a server, if MS goes off and makes mono incompatible with their own version (whether by API changes, implementation secrets or licensing stupidity), chances are developers who use
Well, yes and no. Clocks, battery meters, and other status stuff certainly counts as separate apps, they are being piled on onto PalmOS and such systems as well, just not as cleanly and safely.
Other examples include popping up an on-screen keyboard; or blanking the screen with a clock display when not in use. You still have the issues with multiple windows (in the X sense). Oh, and with a real OS you may well want to have several apps running and switch between them, even though you are only looking at one at a time.
For some of the newer devices, like the Zaurus, the screen is starting to become big enough that it also starts to make sense to actually have windowed applications. The Zaurus has the same resolution as my old laptop, and windowed apps worked fine there.
Oh, while we're on it - I've looked at screenshots from several projects. They generally do the right thing in using the stuff full screen. But then they pile on toolbars, status bars and big honking scrollbars and frames to the point where the actual application area is only half of the total screen estete. Ugh. That is the one thing I find Palm does very well - there is very little wasted screen estate on a palm device.
You know, people have all sorts of reasons for using a given OS other than ideological ones. For example, I'd like a Zaurus with Linux and a GTK-based interface. I would not choose a Zaurus over Palm or iPaq for ideology, but simply because I like the Zaurus form factor a lot better. I would like Linux over PalmOS or WinCE not because of ideology, but because I find it a comfortable and powerful system that I know well. I would like GTK over Qt (ora any of the dozen other toolkits), not because of ideology, but because I again like the GTK programming model, know it fairly well, and have some stuff written with GTK I would enjoy having on a PDA.
Had PalmOS or WinCE offered me a better platform for my needs, I would want them instead.
As for "out of work software engineers" - to be blunt, do I care? Nope. Turn it around: if Palm or anyone else chooses Linux over another system, they do so because it benefits their bottom line. In other words, staying with the previous solution gives the customer a more expensive or less good system. As a customer, I don't really feel the need to sponsor a commercial development team on another continent to the detriment of myself.
To put it even blunter: If they (as a team, not individuals) can't produce value over time at least equal to what they cost the company, they aren't doing the right job in the first place. The company has them doing the wrong job, or the kind of work they are doing is not viable for a commercial entity in the first place.
BTW, what PDA:s does this run on at the moment? My sole reason not to get a zaurus is the lack of GTK.
Something like Tiny X has a total footprint of 700k or thereabouts. X isn't (or doesn't have to be) that resource intensive. Hopefully everybody realizes by now that the "memory hungry" myth is just that; a result of unavoidably disingenoius reporting by system tools. And you don't _need_ a large dynamic font system with AA, or all the modules you load at run-time either, if you want to shrink resource usage further.
And X-less methods rack up that kind of resource use anyway; _something_ has to handle expose and redraw stuff, for instance, if you are aiming for something able to run more than one thing at a time. Apparently (I have not worked on it myself), qtopia requires the applications to handle WM-stuff by themselves.
I agree that Palm did a pretty good thing with their system, by only allowing one - full-screen - application to run at any one time. Easy model to handle, and resource efficient. But as the huge collection of hacks show, it is also quite limiting.
...and only on this site would that detail actually matter...
Gotta love this place.
Um, Open Source community != US-based paid developer community.
The press release specifically mentions plug-ins for Evolution for Novell's server stuff. This is tentatively a very hopeful development.
Well, in the picture, it does have an "enter" key, so I'd say it at least has an optional mode for it.
It's really a matter of perspective.
:(
We have PDA:s that can also make cellular phone calls. We have phones that can double up as PDA:s. They seem to aim for exactly the same market, but, of course, they don't, since they're best features are aimed at different uses.
Same thing with calculators. I'd love to have a HP calculator that will also function reasonably as a PDA. I'm a lot less interested in a PDA that can also do some calculator functions.
It's all about where the focus is. Take the keyboard as an example: a dinky on-screen keyboard, or aphanumeric keyboard just isn't nearly as functional and convenient as a 'real' calculator keyboard a'la my deeply missed HP15, where all the functionality is right there, at your fingertips. Likewise, a phonepad isn't really that good for PDA functionality, and a touch screen isn't really that good for a phone.
Also, the software for PDA:s are of varying, and unknown, quality. One thing that really made the HP line of calculators stand out was their attention to various corner cases. When you got a result, you knew that was the correct one, to the practical limit of the hardware and encoding used. The Palm calculators I've tried have inevitably had various bugs and have missed special cases that made you get the wrong result from time to time - they would not handle over/underflow correctly in all cases, or use algorithms that would not give the stated precision over all of it's range, and so on.
My dream would be a new HP calculator with the format and design of the HP15c, but modernized (faster CPU with more memory; pisel screen, rather than segment, and so on). That one was a nearly perfect unit for me. After fifteen years, I had unfortunately dropped it, spilled coffee and soda in it, buried it under piles of books, stuffed it in dirty, dusty bags and submerged it too many times and it gave up
Even assuming it _is_ just a recompile that's needed (and, especially for OS:es that is far from the case), you still need to convince those holding the source to actually do it. If MS doesn't see a profitable enough market, they will not "recompile for a different platform", and you are sans Windows for your platform. If Oracle doesn't want to bet on the new architecture, you won't have your database available. The same goes for much of the rest of mainstream computing today.
OSS is interesting, as it - like Unix before it - partly changes that equation. With the source out there, a chip maker can do the port themselves, without having to rely on another company to be good to them.
Intel _has_ tried to introduce different platforms in the past - they had a RISC chip going for some years, for example, that never went anywhere. But, no platform ==> no third-party software ==> no system builders to pick it up.
Oh, and both AMD:s and Intel's 64 bit offerings _do_ already provide exactly that backwards bridging you're talking about. So where's the problem?
I went to school in Sweden, yes; I got my master's in CS. I thought briefly about maybe taking a bachelor's in mathematics, but alas, I like math more than math likes me :) I'm just not really focused enough to be able to do good work in the subject.
Multiple choice exams are bad wherever they are used; in a subject like mathematics, it's just ridiculous. In our introductory semester of mathematics, we were 150 students (it was mandatory for those planning to major in CS, physics, mathematics or statistics), and they still managed to have both a written exam and an oral exam for everybody that passed the written one. Granted, there were not nearly 150 people passing the written one, but anyway.
Most math departments have a course somewhere after the introductory sequence which teaches basic proof techniques often by studying the definition of numerical systems from logical axioms.
_After_ the "introductory sequence"? We started right off, in both introductory algebra and calculus I, with the basic proofs in each subject, and the emphasis was all along on understanding the theorems, to the point that the exams are in two parts: one ordinary test, and, for those that pass, a two hour oral exam about the theory.
Not the easiest way to get into the subject, perhaps, but you really do get a very solid theoretical foundation to pin your understanding on.
That said, I don't like calculus... The proof techniques are far too often of the type "half a page of dreary symbol manipulation, apply the magic trick that you would never think of yourself, and another half a page of manipulation". Discrete mathematics (and topology) tends to be a lot, well, cleaner and more beautiful.
There are a lot of issues one can bring up with the test - not identical versions of various software; different X drivers, one distro will have patches missing in the others and so on. Clearly, that greatly influences the results.
And that is a good point to take home. Optimizing compiles is _not_ the panacea for speed and responsiveness that - a minority, I believe - of source-based distros tend to bring up. There are so many other factors intimately involved in it that any benefits are generally lost in the noise.
For some specific components, it can be a good idea - but for those, most distros tend to ship several optimized versions that the installer chooses between at installation time.
Another domain that benefits are specialized, compute-intensive applications; things like simulators or other technical stuff. But then, those apps are generally tweaked and compiled by their users no matter what distro is used anyway.
I don't think that is the kind of problem the previous poster had in mind. Take the example with having a row of tiny buttons with less-than-clear icons on them.
Their functionality - or lack of it - has nothing to do with the problem domain. No matter what the application, they are making it harder to use than would otherwise be the case. They are difficult to see, to distinguish between, and easier to make an accidental "misclick" and select the wrong function.
I fully agree that an UI should be designed for the target user population. A desktop (whether Gnome, KDE, Windows, Mac or whatever) tends to be rather plain an uninnovative, and it must be as the target population is, well, everybody. A smaller, specialized segment, such as Blender users, can motivate a much more specialized, targeted UI as well.
But there are also some things that make using an app harder than it should be, no matter what the target population is (assuming it's at least human, and alive). They are things that stretch the sensory or motor-capabilities of the user too much (like finding the right button, when they are too small and too many, bunched together too tight), or stuff that could be one way or another without loss in functionality, and the "backwards" way is chosen. Those kinds of things would be a pure positive to change, with no downside for anyone.
But then, of course, many potential customers will never log in, opting instead for a different merchant that does publish its prices.
Because the stuff they want to use is available under GPL, but not BSD?
Um, so? I sat on a P2/233 up until this year in the department. With 256Mb memory, running Gnome2 with Mozilla, AbiWord and Evolution was no problem at all. For office/productivity apps, the CPU simply isn't a serious bottleneck; the critical parameter is memory (to avoid swapping).
And why did I sit on such an old machine? Because I was too lazy to get up and order a new one until after New years, since it basically worked fine for what I used it for. Only when it started to act up (hard drive read failure rate was shooting up alarmingly, and the video card was going south) did I feel the need to get a new machine.
It's not about wear and tear. The feel of the thin, flimsy cotton or linen in dress shirts literally makes my skin crawl. If find the feel vaguely repugnant. And no, a thick weave (and I mean a weave with thick strands) of the kind I like is fairly loose, so it lets through as much air (if not more) than the thin - but tight - weave of a dress shirt.
:)
And I have no problem finding shirts of the kind I like that fit otherwise. Sleeve length, width, shoulder play; all doable, except for the collar. The basic problem really is that while dress shirts are sold with various collar and arm sizes, these kind of shirts tend to be one-dimensionally sized - you get bigger collar, you get bigger everything else.
Fortunately, I work in academia, where jeans and t-shirt is perfectly acceptable dress day-to-day dress. Have a shirt, and you're well dressed. Throw on a jacket, and people wonder if I'm due at a wedding or something
I have a problem with shirts. I have a pretty thick neck, so most shirts I look at won't fit. If the neck is wide enough, the arms are far too long. The only recourse would be dress shirts, but, frankly, I hate them. They are only white (=sweat stains galore after a whole day) or baby blue, and are made in, in my taste, way too flimsy material.
I like shirts that use a thick, comfortable, material - heavy unsmoothed linen is fine, for example, or a _really_ heavy cotton weave. I also prefer darker, muted colors. But I seem unable to find such shirts with the neck width I need. Thus I use them, but unbuttoned, and without a tie.
They're just about level with Gamecube, actually, and both are very distant thirds to PS/2. The only market where X-box is going anywhere is the US; sales are not good in Europe, and pretty much disastrous in Asia.
Nah, they're not a monolith. If they were, they'd be in orbit around Jupiter right now instead of producing software. The real test of course comes in 2010 when we find out if MS will eat Jupiter or not.
In some respects .net is more open than Java. This has mostly to do with the core being accpted as a standard, which Sun repeatedly has refused to do with Java.
.net architecture (and observe that I am talking about the standardized core, not the entire class library).
Lately there have been noises that Sun considers _any_ implementation of Java as subject to their licensing and control. This they can do since they own IP and patents on the Java architecture, and since they haven't submitted it as a standard (which would require opening up the licensing). This could not happen with the core
The apps choose what to send cluepackets about. I doubt any app writer would be dumb enough to send out an entered password as a cluepacket, anymore than they would print it in clear text on the screen.
It's a neat idea.
Maybe explaining the reason for it to them? Most people can readily identify with this issue, and framing it as a way to get around it will likely be fairly well received.