You're using the mechanical argument as to how a DLL is loaded. The opposing argument is that the GPL talks about derivative works. If you create a program that has to use a certain DLL (Without a standard interface where you cold drop in a replacement from another source) then the combined program is a derivative work. The FSF goes as far as claiming that your software is a derivative work as well, even though you don't distribute a prelinked version. Notice that this doesn't apply to standard libraries like the C library, but only to specialized libraries.
As I said, it's a gray area, and as I understand it nobody has yet tried to go against the FSF when confronted with this kind arguments, so I'm guessing the lawyers agree that there might be something to it, or that it's not worth the risk.
It's a very rare situation though, and I last saw this a few years ago so maybe it's been decided.
First of all, separate out the LGPL software. It's perfectly OK to distribute these as a library. At most you have to give a written offer for source code, or include a text file with the license. Did you check their "About" dialog? The directory? Even if they're missing that, having the original author contact the company should get them to include this, as it's trivial for them. Just remember, ask for compliance, not money.
GPL'd software is more complicated. the license/offer for source code also applies, but on top of that you have to see if they are linking with the program. If they are calling an executable, they are probably OK. If it's statically linked, they have to GPL their code. If it's a dll it's a gray area. The FSF says their code is derivative, some people argue otherwise. Maybe the threat of a lawsuit would be enough to get them to open their sources but it's not as cut and dried.
Re:EA probably has no interest in making headlines
on
Sexuality And The Sims
·
· Score: 1
Too late. The daily show did a segment on that quite a few years ago. IIRC it was Samantha Bee talking to people who made sexual scenarios, bondage items, etc for the Sims, all the time acting like she didn't get it.
It was pretty funny, but in the end it looked like the Daily Show didn't get it, or hadn't seen the actual mods, until they showed clips of a Sims dungeon as their moment of Zen, showing that they understood what was going on and were playing it for laughs.
Err, after submitting my other post I found this summary of features in Gutsy: As for the new features added in Gutsy Gibbon there are plenty, most of which are particularly well suited for new users to Linux. Some of the more notable new features are a Graphical Configuration tool for X, improvements in plug-in handling for Mozilla Firefox, revamped printing system with PDF printing by default, fast user switching, new desktop search (Tracker) application and the new AppArmor security framework.
Also, for printers: A feature I was particularly impressed with was the new updated printing system found within Ubuntu Gutsy Gibbons. Most printers can now be plugged in and Ubuntu will automatically set up the printer. Also, Ubuntu has implemented a slick new "PDF printer" which is enabled by default. The "PDF printer" allows you to easily created PDF files from all applications such as Gimp, Firefox, Rhythmbox and other non-Gnome applications.
So it looks pretty good. Not much that can't be done already but it seems like a good upgrade, and should be very good for new users.
Hardy's doesn't have features listed yet. From TFA: "Everyone is welcome to think of and develop ideas for features that could be present in the Hardy Heron release." So it'll be a while before there is even a wishlist.
Gutsy's seem to be at the launchpad here. I see: - free-flash - mobile-kernel, mobile-app-framework and related packages, apparently for developing and using the same apps on mobile devices and desktop -xorg7.3 -composite-by-default: Compositing window manager, (but still at drafting stage?) - bullet-proof-x - displayconfig-gtk: "We have got a configuration backend that is currently used in Kubuntu. The GTK frontend supports the basic functionality." - restricted-manager-improvements:" should be able to detect hardware for which the required package is either not installed, or not enabled by default [...] with the above, should be able to offer facilities to help enable the hardware, for example detecting broadcom windows drivers installed alongside or placed on media somewhere, and cutting the firmware out of them" - Virtual machine on CD: "Make it possible to run Ubuntu in a virtual machine when booting the desktop CD under Windows." - write-support-for-ntfs
I don't see much that can't be done in Feisty, but I haven't been using it for long, so I might be missing something. The virtual machine on cd stuff looks very nice for showing people without even a reboot. It looks like improved hardware support, more facilities for the user (screen configuration, etc), some nice defaults (ntfs3g, beryl, etc) and so on.
If someone finds a better description, let me know.
I've said it before and say it again... these silly names are hurting Ubuntu. When you need to convince a boss that installing Ubuntu on office computers is the way to go, you'd need a more professional sounding name. "Windows XP" or "BeOS 5" sounds fine. But "Ubuntu Hardy Heron" does not. Sure you can use the 7.10 number, but it seems that the Ubuntu community prefers not to use the numbers, and these silly names actually crop up within the OS more. What about Mac OS X's Jaguar, Panther, Tiger, Leopard, etc?
Yes, the adjective sounds stupid, but it's not unheard of to have additional names. Just point to the manager's computer next time someone laughs.
Re:It's so simple, but...
on
GWT in Action
·
· Score: 3, Informative
I looked into GWT a month or two ago. From what I understood, you can write both client and server code in Java and then GWT compiles the client to Javascript and the server to a servlet.
You don't actually write any SOAP code, but rather use Java Remoting to call across the wire, getting a Java object back that you can manipulate. The GWT compiler translates that into a XML request and writes all the serialization/deserialization code and so on. In fact it doesn't seem to be able to call an XML web service at all (that's the reason why we didn't pick it up for our UI, our company does business logic in C#).
Since it is also a widget library you should be able to have complex widgets with multiple DOM elements and access them as a unit. You could write your own javascripts objects to access the DOM, but then it's basically a GUI library.
As far as I can tell, the whole point is to write both client and server with a single language and interface, which is very useful.
For example, in our case, the pages are generated with ASP.Net in C#. If we want to disable a button when the page is served we do buttonSubmit.Enabled=false; but if we want to reenable it without a postback, we have to add javascript code to find the DOM node for the button and then enable it. Imagine how messy it gets when you want to add a row to a datagrid. After a while you can't tell all the interactions in the GUI handling logic. I'd love to have something that allows me to write an event handler where I can just write "buttonSubmit.Enabled=true;" and let the compiler work out all the DOM walking code.
So you have three choices: - Serve the full page (formatted in server side language) and then tweak it in javascript, relying on knowledge of the widget library's rendering. - Serve a bare bones page and do all the GUI in Javascript, both initial and updates. Let the server handle business logic only. I see some posters have suggested moving the initial javascript rendering to the server. - Serve a full page and let something like GWT convert the client side code from working with the same objects as the server to a working Javascript/DOM implementation. This is the approach GWT is taking.
Of course, I only looked at it for a couple hours, so someone will probably tell me all the ways I'm wrong.
There is one point people haven't pointed out. You agree to grant Google a non-exclusive license to use your stuff, but can that license be revoked? If I put up an embarrassing video on Google video/Youtube and then take it down, does Google have the right to still use it for promotional purposes or whatever? Do I have any recourse to have them take it down?
Imagine some silly video becoming a meme and you want it to stop. You can ask third party copies to be taken down with the DMCA. But suddenly Google can use it in advertisements for their products because you agreed to it. I wonder if the star wars kid would approve.
Granted, currently they will do no evil and probably would respond favorably if I kindly ask them to stop distributing my video, but what about 5 years down the line?
I'll second that. In addition, without a transplant, she stands a decent chance of living a long full life. Transplant patients don't last that long, on average.
Over 2/3 alive at 5 years, and actually pretty similar at 10 years - bearing in mind that most of bad outcomes are in the first year, and that this is all causes of death, including deaths that were unrelated to the transplant. I think you're missing the point here. A 15 year old living 10 more years would mean she dies at 25. Not what I would call a "long full life".
I have to agree with the grandparent, this looks very promising. I bet this clinical trial just got extended to a lot of other conditions. If they can generalize this to help hearts come back from a variety of heart problems at different ages we're talking about Nobel prize quality research.
And it's up to you whether to assign copyright or fork the software.
The only place I've heard that do that is the FSF, and I'd happily assign my copyrights to them as I know they are going to do the right thing. On the other hand, assigning copyrights to a private company? Forget it. They can't be trusted to put morals before profit.
As Izchak says, "Slow down. Think clearly.". In Nethack you can stop, look at your inventory, and think of the best way out, and there is often one or many. Your chances to survive drop if you're full of adrenaline and in fight or flight mode. I know, I've had it happen many times.
This also applies to most horror games, but it ruins the mood. You have to decide if you're playing to win or to "enjoy" it.
I don't know who modded this up, but the question doesn't make any sense. People who make GPL software aren't bound by the GPL with regards to their own software. At least until the first time they accept a patch. At that point they have to abide by the patch author's license, i.e. the GPL.
And let's face it, who distributes GPL'd software and doesn't accept patches?
But we're safe. The FSF has defused this threat. The bullies are afraid to take these issues to court. No need to be worried about technical violations. The spirit of the GPL is most certainly violated by such schemes. Yes, even the spirit of GPL 2. Courts do look at intent when considering cases. But good to have it spelled out in GPL 3, to avoid confusion. Err. Wrong.
A patent license is "discriminatory" if it does not include within the scope of its coverage, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the rights that are specifically granted under this License. You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license (a) in connection with copies of the covered work conveyed by you (or copies made from those copies), or (b) primarily for and in connection with specific products or compilations that contain the covered work, unless you entered into that arrangement, or that patent license was granted, prior to 28 March 2007.
That specific case was considered during the drafting process and questions went out to the community asking whether the new clause should apply retroactively to the Novell deal. The consensus was to exclude previous agreements, including MS/Novell's.
Sandros and Linspire on the other hand, are screwed.
Now if this keeps up I wonder if Roland will: -Keep submitting stories. (Reputation is good for business) -Go away. (No need to waste his time) -Send a C&D to slashdot and subsequently sue. (Editing his submission is clearly the creation of a derivative work and needs an additional permission from him to be distributed).
Time for Taco to add an EULA where you state that your submission might be mercilessly edited.
[In court] Okay, so what am I doing? [Sees other lawyer in court] I'm prosecuting this guy. [Man makes opening statement] Nope. He's prosecuting me.
Tough another quote sticks with me on this case: "Facts, not memories. That's how you investigate. I know, it's what I used to do."
Leonard Shelby. Memento
It's in that same spirit that I'm voting Republican in the next presidential election. Do you REALLY think one party rule is going to better under Democrats? Err... Is there anything stopping you from voting democrat for president and republican for congress on the same election? Is that somehow not possible in your state?
If no one in the building but HR and your line report need to know your name, you're doing your job... And that's easy to quantify. Number of outages per week, average downtime, etc. Then report this by service (the email server was up 100% of the time, but the server with lousy intranet app X crashed twice, they need to rewrite that).
Of course, it's not productivity, but it's a measurement of the quality of service. Combine that with other indicators like users served, requests serviced, emails delivered, etc. and you can actually chart improvements in "productivity".
Even something like average time to solve a ticket or bring up a server is a useful indicator. Granted, it'll vary depending on what the failure is, but over time it should average out to a useful picture.
There are three aspects to system administration: Making sure everything keeps working, expansion and support.
For maintenance you'd want to look at simple downtime measurements. Then you can talk about schedule/unscheduled, etc. If you already have monitoring systems in place then this is relatively simple.
For expansion, I suggest you include any new expansion as a small "project", with a due date. Then you report %projects completed on time. The danger is having your boss then require a 10 page proposal, approval from 5 different people and a budget estimate, etc. But on most expansions that require nontrivial hardware you probably do at least the budget anyway.
For support (creating new users, fixing things, help desk stuff) you'd want to look at tickets/day and to use satisfaction surveys (automatically emailed?)
In fact, you might want to consider surveys as a regular tool. If you can show that most people consider the servers to be stable, then that's something. Of course, from the self-selected nature of these surveys, most people who fill them will have something bad to say about at least one aspect, so you have to keep that in mind
Most of these things are a good idea to record anyway, even if your boss isn't a bean-counter. Looking at trends and setting reasonable goals for your team is a reasonable way to detect problems and encourage improvement.
Well, the point of trying to prove or falsify physical theories isn't so much to prove it but rather to differentiate among the different theories.
Basically you need a situation where the outcomes predicted by string theory, M-theory, relativity, quantum gravity, etc. are different. Then you run an experiment emulating this situation and see which one is right. This doesn't prove a theory but narrows down the possibilities.
If I understand it correctly, right now the problem is that modern theories all explain current observations equally well, and they have become complicated enough that it's hard to come up with appropriate experiments. Additionally, the experiments people can come up with are not feasible to perform.
The good news is that since we can't currently build anything that behaves differently depending on which theory is right, then there's no hurry to figure it out, as it won't affect our engineering capabilities for some time. At least not until someone does come up with a process that works under a specific theory (like nuclear power does) or our capabilities increase (like GPS, that requires compensating for relativity).
I wonder about Stephen Hawking's software. IIRC he selects the beginning of a word (by timing) and then a menu of possible completions are shown.
Though maybe the patent examiner did look at these things (tab completion, intellisense, interfaces for the disabled) since they are relatively well known, and only granted the claims for new things.
Slashdot Mirror
In theory yes, but then you'd have to code a whole dll that does what the original does, at which point why use the original?
I'm not saying it's not technically possible or even a sound argument. I'm saying: would you bet the courts will see it your way?
You're using the mechanical argument as to how a DLL is loaded. The opposing argument is that the GPL talks about derivative works. If you create a program that has to use a certain DLL (Without a standard interface where you cold drop in a replacement from another source) then the combined program is a derivative work. The FSF goes as far as claiming that your software is a derivative work as well, even though you don't distribute a prelinked version. Notice that this doesn't apply to standard libraries like the C library, but only to specialized libraries.
As I said, it's a gray area, and as I understand it nobody has yet tried to go against the FSF when confronted with this kind arguments, so I'm guessing the lawyers agree that there might be something to it, or that it's not worth the risk.
It's a very rare situation though, and I last saw this a few years ago so maybe it's been decided.
Or maybe track the phone call?
(yeah yeah, everyone wants to be James Bond and doesn't care for old-school solutions)
First of all, separate out the LGPL software. It's perfectly OK to distribute these as a library. At most you have to give a written offer for source code, or include a text file with the license. Did you check their "About" dialog? The directory? Even if they're missing that, having the original author contact the company should get them to include this, as it's trivial for them. Just remember, ask for compliance, not money.
GPL'd software is more complicated. the license/offer for source code also applies, but on top of that you have to see if they are linking with the program. If they are calling an executable, they are probably OK. If it's statically linked, they have to GPL their code. If it's a dll it's a gray area. The FSF says their code is derivative, some people argue otherwise. Maybe the threat of a lawsuit would be enough to get them to open their sources but it's not as cut and dried.
Too late. The daily show did a segment on that quite a few years ago. IIRC it was Samantha Bee talking to people who made sexual scenarios, bondage items, etc for the Sims, all the time acting like she didn't get it.
It was pretty funny, but in the end it looked like the Daily Show didn't get it, or hadn't seen the actual mods, until they showed clips of a Sims dungeon as their moment of Zen, showing that they understood what was going on and were playing it for laughs.
Err, after submitting my other post I found this summary of features in Gutsy:
As for the new features added in Gutsy Gibbon there are plenty, most of which are particularly well suited for new users to Linux. Some of the more notable new features are a Graphical Configuration tool for X, improvements in plug-in handling for Mozilla Firefox, revamped printing system with PDF printing by default, fast user switching, new desktop search (Tracker) application and the new AppArmor security framework.
Also, for printers:
A feature I was particularly impressed with was the new updated printing system found within Ubuntu Gutsy Gibbons. Most printers can now be plugged in and Ubuntu will automatically set up the printer. Also, Ubuntu has implemented a slick new "PDF printer" which is enabled by default. The "PDF printer" allows you to easily created PDF files from all applications such as Gimp, Firefox, Rhythmbox and other non-Gnome applications.
So it looks pretty good. Not much that can't be done already but it seems like a good upgrade, and should be very good for new users.
As for Hardy, they're still taking proposals.
Hardy's doesn't have features listed yet. From TFA: "Everyone is welcome to think of and develop ideas for features that could be present in the Hardy Heron release." So it'll be a while before there is even a wishlist.
Gutsy's seem to be at the launchpad here. I see:
- free-flash
- mobile-kernel, mobile-app-framework and related packages, apparently for developing and using the same apps on mobile devices and desktop
-xorg7.3
-composite-by-default: Compositing window manager, (but still at drafting stage?)
- bullet-proof-x
- displayconfig-gtk: "We have got a configuration backend that is currently used in Kubuntu. The GTK frontend supports the basic functionality."
- restricted-manager-improvements:" should be able to detect hardware for which the required package is either not installed, or not enabled by default [...] with the above, should be able to offer facilities to help enable the hardware, for example detecting broadcom windows drivers installed alongside or placed on media somewhere, and cutting the firmware out of them"
- Virtual machine on CD: "Make it possible to run Ubuntu in a virtual machine when booting the desktop CD under Windows."
- write-support-for-ntfs
I don't see much that can't be done in Feisty, but I haven't been using it for long, so I might be missing something. The virtual machine on cd stuff looks very nice for showing people without even a reboot. It looks like improved hardware support, more facilities for the user (screen configuration, etc), some nice defaults (ntfs3g, beryl, etc) and so on.
If someone finds a better description, let me know.
Yes, the adjective sounds stupid, but it's not unheard of to have additional names. Just point to the manager's computer next time someone laughs.
I looked into GWT a month or two ago. From what I understood, you can write both client and server code in Java and then GWT compiles the client to Javascript and the server to a servlet.
You don't actually write any SOAP code, but rather use Java Remoting to call across the wire, getting a Java object back that you can manipulate. The GWT compiler translates that into a XML request and writes all the serialization/deserialization code and so on. In fact it doesn't seem to be able to call an XML web service at all (that's the reason why we didn't pick it up for our UI, our company does business logic in C#).
Since it is also a widget library you should be able to have complex widgets with multiple DOM elements and access them as a unit. You could write your own javascripts objects to access the DOM, but then it's basically a GUI library.
As far as I can tell, the whole point is to write both client and server with a single language and interface, which is very useful.
For example, in our case, the pages are generated with ASP.Net in C#. If we want to disable a button when the page is served we do buttonSubmit.Enabled=false; but if we want to reenable it without a postback, we have to add javascript code to find the DOM node for the button and then enable it. Imagine how messy it gets when you want to add a row to a datagrid. After a while you can't tell all the interactions in the GUI handling logic. I'd love to have something that allows me to write an event handler where I can just write "buttonSubmit.Enabled=true;" and let the compiler work out all the DOM walking code.
So you have three choices:
- Serve the full page (formatted in server side language) and then tweak it in javascript, relying on knowledge of the widget library's rendering.
- Serve a bare bones page and do all the GUI in Javascript, both initial and updates. Let the server handle business logic only. I see some posters have suggested moving the initial javascript rendering to the server.
- Serve a full page and let something like GWT convert the client side code from working with the same objects as the server to a working Javascript/DOM implementation. This is the approach GWT is taking.
Of course, I only looked at it for a couple hours, so someone will probably tell me all the ways I'm wrong.
There is one point people haven't pointed out. You agree to grant Google a non-exclusive license to use your stuff, but can that license be revoked? If I put up an embarrassing video on Google video/Youtube and then take it down, does Google have the right to still use it for promotional purposes or whatever? Do I have any recourse to have them take it down?
Imagine some silly video becoming a meme and you want it to stop. You can ask third party copies to be taken down with the DMCA. But suddenly Google can use it in advertisements for their products because you agreed to it. I wonder if the star wars kid would approve.
Granted, currently they will do no evil and probably would respond favorably if I kindly ask them to stop distributing my video, but what about 5 years down the line?
Over 2/3 alive at 5 years, and actually pretty similar at 10 years - bearing in mind that most of bad outcomes are in the first year, and that this is all causes of death, including deaths that were unrelated to the transplant. I think you're missing the point here. A 15 year old living 10 more years would mean she dies at 25. Not what I would call a "long full life".
I have to agree with the grandparent, this looks very promising. I bet this clinical trial just got extended to a lot of other conditions. If they can generalize this to help hearts come back from a variety of heart problems at different ages we're talking about Nobel prize quality research.
And it's up to you whether to assign copyright or fork the software.
The only place I've heard that do that is the FSF, and I'd happily assign my copyrights to them as I know they are going to do the right thing. On the other hand, assigning copyrights to a private company? Forget it. They can't be trusted to put morals before profit.
As Izchak says, "Slow down. Think clearly.". In Nethack you can stop, look at your inventory, and think of the best way out, and there is often one or many. Your chances to survive drop if you're full of adrenaline and in fight or flight mode. I know, I've had it happen many times.
This also applies to most horror games, but it ruins the mood. You have to decide if you're playing to win or to "enjoy" it.
And let's face it, who distributes GPL'd software and doesn't accept patches?
A patent license is "discriminatory" if it does not include within the scope of its coverage, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the rights that are specifically granted under this License. You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license (a) in connection with copies of the covered work conveyed by you (or copies made from those copies), or (b) primarily for and in connection with specific products or compilations that contain the covered work, unless you entered into that arrangement, or that patent license was granted, prior to 28 March 2007.
That specific case was considered during the drafting process and questions went out to the community asking whether the new clause should apply retroactively to the Novell deal. The consensus was to exclude previous agreements, including MS/Novell's.
Sandros and Linspire on the other hand, are screwed.
So, assuming this is a .5kg prop, it would mean they paid $4500 for a lot of publicity and goodwill. Seems like a great deal to me.
If congress dislikes it, they can set up a donation drive, I'm sure they'll get more than $4500 if they agree to put it on a museum afterwards.
BTW, Every shuttle mission carries certain memorabilia. See for example the challenger flag which survided the Challenger accident.
Now if this keeps up I wonder if Roland will:
-Keep submitting stories. (Reputation is good for business)
-Go away. (No need to waste his time)
-Send a C&D to slashdot and subsequently sue. (Editing his submission is clearly the creation of a derivative work and needs an additional permission from him to be distributed).
Time for Taco to add an EULA where you state that your submission might be mercilessly edited.
[In court] Okay, so what am I doing?
[Sees other lawyer in court] I'm prosecuting this guy.
[Man makes opening statement] Nope. He's prosecuting me.
Tough another quote sticks with me on this case:
"Facts, not memories. That's how you investigate. I know, it's what I used to do."
Leonard Shelby. Memento
Makes me wonder if there is any way congress can get one of these seven appointed dismissed attorneys as US Attorney General.
Then again, if Bush manages a deal good for his administration and posts one of them, that might be a nice PR coup.
Of course, it's not productivity, but it's a measurement of the quality of service. Combine that with other indicators like users served, requests serviced, emails delivered, etc. and you can actually chart improvements in "productivity".
Even something like average time to solve a ticket or bring up a server is a useful indicator. Granted, it'll vary depending on what the failure is, but over time it should average out to a useful picture.
There are three aspects to system administration: Making sure everything keeps working, expansion and support.
For maintenance you'd want to look at simple downtime measurements. Then you can talk about schedule/unscheduled, etc. If you already have monitoring systems in place then this is relatively simple.
For expansion, I suggest you include any new expansion as a small "project", with a due date. Then you report %projects completed on time. The danger is having your boss then require a 10 page proposal, approval from 5 different people and a budget estimate, etc. But on most expansions that require nontrivial hardware you probably do at least the budget anyway.
For support (creating new users, fixing things, help desk stuff) you'd want to look at tickets/day and to use satisfaction surveys (automatically emailed?)
In fact, you might want to consider surveys as a regular tool. If you can show that most people consider the servers to be stable, then that's something. Of course, from the self-selected nature of these surveys, most people who fill them will have something bad to say about at least one aspect, so you have to keep that in mind
Most of these things are a good idea to record anyway, even if your boss isn't a bean-counter. Looking at trends and setting reasonable goals for your team is a reasonable way to detect problems and encourage improvement.
Well, the point of trying to prove or falsify physical theories isn't so much to prove it but rather to differentiate among the different theories.
Basically you need a situation where the outcomes predicted by string theory, M-theory, relativity, quantum gravity, etc. are different. Then you run an experiment emulating this situation and see which one is right. This doesn't prove a theory but narrows down the possibilities.
If I understand it correctly, right now the problem is that modern theories all explain current observations equally well, and they have become complicated enough that it's hard to come up with appropriate experiments. Additionally, the experiments people can come up with are not feasible to perform.
The good news is that since we can't currently build anything that behaves differently depending on which theory is right, then there's no hurry to figure it out, as it won't affect our engineering capabilities for some time. At least not until someone does come up with a process that works under a specific theory (like nuclear power does) or our capabilities increase (like GPS, that requires compensating for relativity).
I wonder about Stephen Hawking's software. IIRC he selects the beginning of a word (by timing) and then a menu of possible completions are shown.
Though maybe the patent examiner did look at these things (tab completion, intellisense, interfaces for the disabled) since they are relatively well known, and only granted the claims for new things.
They'll probably try to sue her anyway.
The big question is: Will she see it coming?