Valve has always said that if they were to go out of business, or shutdown their Steam service, they would release patches to all the games to make them playable without requiring Steam.
They may have said that, but unless it's actually part of their terms and conditions they may find themselves unable to actually do it. When a company reaches the point where its officers realise that it is likely to enter bankruptcy, it becomes illegal to dispose of company assets in a way that does not provide an adequate return for either investors or creditors. If Valve faced bankruptcy, they would legally be required to keep Steam in a position where it could be sold off to a competitor until after they were no longer in control of it.
That's Microsoft's store layout trademark they registered in 2011.
Samething, minus the glass out front.
No, it isn't. MS's layout has curved tables along the sides of the store, and lacks both the table and video screens on the rear wall. Also, their lighting appears to protrude from the ceiling while Apple's is integrated into the ceiling. There's also an additional middle row of tables that would possibly prevent it from being infringing even by itself.
I'm all for living within our current energy means in a reasonable way (and I abhor the pollution from mining and burning coal and oil), but she cites a calculation that projects exponential growth on Earth forward a few hundred years, calculates we will need to cover the whole Earth in solar panels (and then the Galaxy), and then concludes from that somehow that we should stay the way we are.
Not just that -- that calculation explicitly states this: "The purpose of this exploration is to point out the absurdity that results from the assumption that we can continue growing our use of energy." She then goes on to use it as a basis of saying that because our energy use is going to carry on growing, we can't use solar power... she basically missed the entire point of the article. Yes, solar power can't continue providing for growing energy use at the current rate for more than a couple of hundred years. But neither can any other technology -- within 400 years, those same calculations and the assumption of using a 100%-thermodynamically efficient process to produce power from some stored energy reserve put the temperature of the Earth's surface at a level that would wipe humanity out (i.e., average surface temperature exceeds 60C).
The store features a clear glass storefront surrounded by a paneled facade consisting of large, rectangular horizontal panels over the top of the glass front, and two narrower panels stacked on either side of the storefront
The picture you link to shows a store whose storefront has the horizontal panel but AFAICT is lacking the narrower side panels described. Also, without being able to see inside the store, it is impossible to tell if the arrangement of lighting, tables, seating, shelves and video panels all conform to the description provided. It is quite likely that they do not, as Apple's description is very specific in certain details. Note that all of the details would need to match for the store to be considered infringing.
The summary makes this sound much worse than it actually is. To be infringing, a store would have to:
1. Have the same arrangement of windows (not just the single large panel mentioned in the summary, but also the smaller side panels), *and* 2. Have cantilevered shelves, *and* 3. Have multiple rectangular tables, *and* 4. Have flush-mounted video screens on the rear wall.
It's possible a court *might* hold that something hitting 3 out of 4 of these was confusingly similar, but by no means certain. Courts aren't stupid.
Matters of parent companies aren't really the issue; the only relevant question is where does the copyright actually reside, i.e. on which company's balance sheet is it shown -- a question which should be possible to resolve by looking at relevant SEC filings. My guess is most media companies hold their copyrights in a standard US company. Most countries (the US included) only recognize copyright of a foreign-owned work for as long as its owner's country also recognizes it. US copyright lasts longer than anyone else's.
You assume the main load requirements are in sending the notifications of events to other players, but it is more likely that most of the load is consumed by calculating the consequences of each action. This will involve interactions with other players, but these can probably be optimized based on the precise location of the player's ship rather than considering all other players in the system, meaning that I suspect the overall system is behaving more like O(n log n) than O(n^2). (obviously as number of players continues to increase that n^2 term would eventually take over as the most important, but I suspect that's at quite a few more than 3000 players).
So all this time when games I played dropped to 5-10 fps, it wan't due to unoptimized code, crappy drivers, or old hardware. It was actually cutting-edge technology called Time Dilation!
Actually, this time dilation thing is a solution to that very problem. If you are forced to accept that your game is sometimes going to drop to 5fps, basically because you have no way of controlling how much work is necessary for a frame to be calculated, then you need to find a way of working around it. Typically a game is unplayable at that frame rate because too much stuff happens between each frame -- you'll be perfectly fine in one frame, but by the time the next one comes around you've been shot several times, and it'll be two entire frames before you can do anything about it. What EVE does is simply reduce the amount of stuff that happens per frame to compensate, much like if an FPS slowed its physics engine down rather than trying to carry on in real time as they usually do.
2. The core of EVE online was written in the late 90's. Multicores were not really on the radar back then. They became practical and affordable only years later.
Everyone paying attention has known that multiprocessing was the future since, at the latest, the early 90s. Here's a quote from a 1993 research paper:
Resizing will continue to draw applications off mainframes onto networked PCs, workstations, and small servers; simple multiprocessors; large scalable multiprocessors; and massively parallel processing multicomputers. This trend will accelerate as network cost, performance, and maintainability improve.
Simple multiprocessors will be the mainline of computing throughout this decade, to be replaced eventually by various forms of highly parallel processing computers.
This was a typical conclusion of the (many) discussions of the era. Yes, multicore was a bit of a surprise, but servers with slots for multiple single-core processors were commonplace when EVE launched. It was practically a standard feature of most high-end server systems, so writing their servers not to take advantage of such systems was short-sighted. Also several RISC processor designs had emerged that used multiple threads to offset the delay between an instruction being executed and its results being available for a subsequent instruction to use, and to minimize latency due to branches. Everyone knew the future of high performance computing was in multiprocessing and multithreading.
Granted, the inability of CPython to run truly concurrent threads would deter me from using it for applications requiring high performance. But 10 years ago I simply wouldn't have lost a single thought about concurrent threads. Would you?
Yes. I stopped using python circa 2002 (dated by the fact that I evaluated a new-at-the-time system called 'psyco' for improving python performance just before giving up on it, and the earliest-dated reference I can find to this system is from october 2002) when I realised that this problem was not going to be fixed and would become critically important for any large-scale application with non-trivial interactions between users. I'm pretty sure from discussions I saw on message boards and mailing lists at the time that I'm far from the only one who gave up on it.
TFA states there were ~$25k in loss and damages. That is $25k that people paid EVE for (virtual) merchandise that they don't have anymore.
No, it isn't. $25k is what they'd have to pay if they wanted to convert cash to in-game currency in order to recoup their losses, but most of them probably acquired that merchandise through in-game actions, and will likely recover their losses through in-game actions too, so no real currency will actually be involved for a very large proportion of that loss.
The only sense in which it's not "legal" to trade ISK for dollars is that it's against the rules of the game and CCP can ban you from the game if they catch you doing it. There's no actual real-world law against it.
Actually, there quite possibly is. I looked into the possibility of running a game where such trades would be allowed a few years back, and it turns that anti-money-laundering legislation prevents it in at least some countries.
It's an improvement on the old situation because events were timed in real time even with bad server lag, so if you have a 30-second cooldown on some ability or other, but a 5 minute lag, you only get to use it one tenth as many times as you're supposed to, which breaks the game design and is really infuriating to play. Now they slow the cooldowns down along with the lag in responding to commands so you can still keep doing stuff as much as the designers intended you to be able to. Kind of like the difference between the whole game slowing down and just dropping frames to compensate for lag in an FPS -- dropping frames is the easiest design solution, but if it happens too much it'll just get you killed. Slowing the entire game down lets you carry on playing even when the lag's really bad.
Hell, with the prodicts you mentioned you're not even ALLOWED to try to figure out how they work. That's called reverse engineering and it's against the license.
Here in the UK, the right to reverse engineer is legally mandated by statute, so they can't take it away with license terms.
It is not an infringement of copyright for a lawful user of a copy of a computer program to observe, study or test the functioning of the program in order to determine the ideas and principles which underlie any element of the program if he does so while performing any of the acts of loading, displaying, running, transmitting or storing the program which he is entitled to do.
and
Article 6 Decompilation 1. The authorisation of the rightholder shall not be required where reproduction of the code and translation of its form within the meaning of points (a) and (b) of Article 4(1) are indispensable to obtain the information necessary to achieve the interoperability of an independently created computer program with other programs, provided that the following conditions are met: (a) those acts are performed by the licensee or by another person having a right to use a copy of a program, or on their behalf by a person authorised to do so; (b) the information necessary to achieve interoperability has not previously been readily available to the persons referred to in point (a); and (c) those acts are confined to the parts of the original program which are necessary in order to achieve interoperability.
I was thinking along the same lines. Kudos and attaboys to Google for doing this, but what this gift unintentionally did was cause the schools to need to buy I/O devices for those machines. Why not 15000 OLPCs, which are 'complete'?
Because OLPCs have a different focus. OLPC is intended as a general purpose learning tool, while RPi is intended specifically to teach electronics & computer science. Presumably goolgle wanted the latter outcome rather than the former.
While I do want a Simcity game, as SimCity 4 is 9 years old at this point, I can't see this new product as a Simcity game. There is no local saves. [...] At this rate, I doubt it will have any fun cheat codes as was a staple of the series.
Cheat codes? Who needs 'em? I always used to edit the save file and give myself a virtually unlimited budget. Guess that one's gone, too...
To be fair on OP, it seems that at least some of the applications they want to run are running under Windows and/or Mac environments, for which this solution does not apply.
Virtualizing a single application's windows from a remote machine is a non-trivial task that AFAIK hasn't been implemented in open source software for either of these platforms. The closest you'll get is by virtualizing the entire OS -- VirtualBox with Windows guests (and Windows only) can do this. You'd then have to run the virtualbox virtual machine process as an X client, and use X-over-SSH forwarding as described in many existing posts to get the windows to appear on the target machine. Performance will be poor (although my one experience of citrix suggests its performance was equally bad, so maybe you can tolerate that).
Given that the desktops are almost certainly running Windows, this by itself is only half of the solution. The other half is cygwin's X server with the -rootless or -multiwin command line options (the former requires you to run a native X11 window manager, the latter uses Windows' desktop window manager to provide window decorations).
Whilst individual islamic groups may do these things, they are not usually supported by the official churches of their religion. Most Muslims condemn such actions as not in accord with their actual religious beliefs. Scientology has made this an official policy, and abuse like this goes to the highest level. You'd find it hard to find actual practicing scientologists who would speak out against such acts.
Arduino from any other AVR board. It is a C-like language with a library; you're not programming to the bare board, you're not even writing your own main() routine. It is not intended for profressional programers, the target audience appears to be "multidisciplinary" (ie, people who aren't programmers).
I've been programming professionally for fifteen years - in C, Perl, PHP, Javascript, VB6, Actionscript, and other languages. Being a programmer, I was glad I didn't have to learn both embedded systems and assembler at the same time. For a guy like me, at a point where I've done just a little bit of kernel ptogramming for example, Arduino was really nice. "mov 0x40 0xD0" isn't what most programmers are familiar with.
I've been working directly with AVR chips rather than an Arduino on several projects over the last 6 months, and have never once had to touch an assembler. C is perfectly adequate for everything you're likely to want to do.
Raspberry Pi and Beaglebone both run Linux. The RPi is super cheap but is better targeted at apps which require a GUI. The BeagleBoard is more expensive but is better tuned for embedded use. It would be nice if the inverse were true, but oh well.
The Olimex A13-OLinuxIno is a reasonable compromise -- priced between the two boards, but much closer to the feature set of the beagleboard (and actually a faster processor with more memory than either). It was intended as a Linux-running competitor for Arduino, and is pretty nifty, as long as you're happy with the fact that it runs on a cheap chinese SoC processor that is only officially supported as a platform for making low-end Android tablets, so the Ubuntu port is not officially supported. And like Arduino, it's an open-hardware design. The only downside as far as I'm concerned is that it doesn't support either DVI or HDMI -- the only way of connecting a display is either via the integrated LCD panel port or on VGA, and resolution is limited to 800x600. But those constraints are fine for embedded use.
I wish to ${deity} that routers had a "reverse https proxy" function that would accept inbound https connections, strip the ssl, and transparently forward the traffic to the same port of an internal IP address where there's a device that's too stupid to know how to do SSL.
Bullshit. If your device has a reason to create an outbound connection, it is (for the most part) limited to one connection to one place for a specific purpose. (Disregarding intentionally buggered on-board software designed with malicious intent).
You're disregarding exactly the situation the GGP post was describing as the reason he turned UPNP off. GP's reply was a reasonable response: if you're assuming that software inside your network is malicious, it doesn't need UPNP to cause mischief... it'll probably hook up to an IRC server or similar in order to accept incoming commands, so that isn't a good reason to disable UPNP.
Now, this situation is (presumably) not malicious, but that doesn't make GP's response invalid. OTOH, I have to query how rare situations like this are. Very few devices automatically create firewall holes for themselves without user confirmation. Most UPNP routers make it very easy to monitor what holes you do have. The proportion of such devices that have massive security flaws like this is also likely to be low. I'm not therefore convinced that this situation is, on balance, enough to make me want to turn UPNP off.
Slashdot Mirror
Valve has always said that if they were to go out of business, or shutdown their Steam service, they would release patches to all the games to make them playable without requiring Steam.
They may have said that, but unless it's actually part of their terms and conditions they may find themselves unable to actually do it. When a company reaches the point where its officers realise that it is likely to enter bankruptcy, it becomes illegal to dispose of company assets in a way that does not provide an adequate return for either investors or creditors. If Valve faced bankruptcy, they would legally be required to keep Steam in a position where it could be sold off to a competitor until after they were no longer in control of it.
Something like this?
http://tsdr.uspto.gov/documentviewer?caseId=sn85194406&docId=DRW20101213072755
That's Microsoft's store layout trademark they registered in 2011.
Samething, minus the glass out front.
No, it isn't. MS's layout has curved tables along the sides of the store, and lacks both the table and video screens on the rear wall. Also, their lighting appears to protrude from the ceiling while Apple's is integrated into the ceiling. There's also an additional middle row of tables that would possibly prevent it from being infringing even by itself.
I'm all for living within our current energy means in a reasonable way (and I abhor the pollution from mining and burning coal and oil), but she cites a calculation that projects exponential growth on Earth forward a few hundred years, calculates we will need to cover the whole Earth in solar panels (and then the Galaxy), and then concludes from that somehow that we should stay the way we are.
Not just that -- that calculation explicitly states this: "The purpose of this exploration is to point out the absurdity that results from the assumption that we can continue growing our use of energy." She then goes on to use it as a basis of saying that because our energy use is going to carry on growing, we can't use solar power... she basically missed the entire point of the article. Yes, solar power can't continue providing for growing energy use at the current rate for more than a couple of hundred years. But neither can any other technology -- within 400 years, those same calculations and the assumption of using a 100%-thermodynamically efficient process to produce power from some stored energy reserve put the temperature of the Earth's surface at a level that would wipe humanity out (i.e., average surface temperature exceeds 60C).
Maybe I don't care if my code rots in obscurity.
I hereby release this post into the public domain.
If you don't care, why release it on a public web site? Just storing it on your local system is easier.
that may be true but can you explain to me how this windows store would not infringe
pricture of a windows store
regards
John Jones
From the textual description of the mark:
The picture you link to shows a store whose storefront has the horizontal panel but AFAICT is lacking the narrower side panels described. Also, without being able to see inside the store, it is impossible to tell if the arrangement of lighting, tables, seating, shelves and video panels all conform to the description provided. It is quite likely that they do not, as Apple's description is very specific in certain details. Note that all of the details would need to match for the store to be considered infringing.
The summary makes this sound much worse than it actually is. To be infringing, a store would have to:
1. Have the same arrangement of windows (not just the single large panel mentioned in the summary, but also the smaller side panels), *and*
2. Have cantilevered shelves, *and*
3. Have multiple rectangular tables, *and*
4. Have flush-mounted video screens on the rear wall.
It's possible a court *might* hold that something hitting 3 out of 4 of these was confusingly similar, but by no means certain. Courts aren't stupid.
Matters of parent companies aren't really the issue; the only relevant question is where does the copyright actually reside, i.e. on which company's balance sheet is it shown -- a question which should be possible to resolve by looking at relevant SEC filings. My guess is most media companies hold their copyrights in a standard US company. Most countries (the US included) only recognize copyright of a foreign-owned work for as long as its owner's country also recognizes it. US copyright lasts longer than anyone else's.
You assume the main load requirements are in sending the notifications of events to other players, but it is more likely that most of the load is consumed by calculating the consequences of each action. This will involve interactions with other players, but these can probably be optimized based on the precise location of the player's ship rather than considering all other players in the system, meaning that I suspect the overall system is behaving more like O(n log n) than O(n^2). (obviously as number of players continues to increase that n^2 term would eventually take over as the most important, but I suspect that's at quite a few more than 3000 players).
So all this time when games I played dropped to 5-10 fps, it wan't due to unoptimized code, crappy drivers, or old hardware. It was actually cutting-edge technology called Time Dilation!
Actually, this time dilation thing is a solution to that very problem. If you are forced to accept that your game is sometimes going to drop to 5fps, basically because you have no way of controlling how much work is necessary for a frame to be calculated, then you need to find a way of working around it. Typically a game is unplayable at that frame rate because too much stuff happens between each frame -- you'll be perfectly fine in one frame, but by the time the next one comes around you've been shot several times, and it'll be two entire frames before you can do anything about it. What EVE does is simply reduce the amount of stuff that happens per frame to compensate, much like if an FPS slowed its physics engine down rather than trying to carry on in real time as they usually do.
2. The core of EVE online was written in the late 90's. Multicores were not really on the radar back then. They became practical and affordable only years later.
Everyone paying attention has known that multiprocessing was the future since, at the latest, the early 90s. Here's a quote from a 1993 research paper:
This was a typical conclusion of the (many) discussions of the era. Yes, multicore was a bit of a surprise, but servers with slots for multiple single-core processors were commonplace when EVE launched. It was practically a standard feature of most high-end server systems, so writing their servers not to take advantage of such systems was short-sighted. Also several RISC processor designs had emerged that used multiple threads to offset the delay between an instruction being executed and its results being available for a subsequent instruction to use, and to minimize latency due to branches. Everyone knew the future of high performance computing was in multiprocessing and multithreading.
Granted, the inability of CPython to run truly concurrent threads would deter me from using it for applications requiring high performance. But 10 years ago I simply wouldn't have lost a single thought about concurrent threads. Would you?
Yes. I stopped using python circa 2002 (dated by the fact that I evaluated a new-at-the-time system called 'psyco' for improving python performance just before giving up on it, and the earliest-dated reference I can find to this system is from october 2002) when I realised that this problem was not going to be fixed and would become critically important for any large-scale application with non-trivial interactions between users. I'm pretty sure from discussions I saw on message boards and mailing lists at the time that I'm far from the only one who gave up on it.
TFA states there were ~$25k in loss and damages. That is $25k that people paid EVE for (virtual) merchandise that they don't have anymore.
No, it isn't. $25k is what they'd have to pay if they wanted to convert cash to in-game currency in order to recoup their losses, but most of them probably acquired that merchandise through in-game actions, and will likely recover their losses through in-game actions too, so no real currency will actually be involved for a very large proportion of that loss.
The only sense in which it's not "legal" to trade ISK for dollars is that it's against the rules of the game and CCP can ban you from the game if they catch you doing it. There's no actual real-world law against it.
Actually, there quite possibly is. I looked into the possibility of running a game where such trades would be allowed a few years back, and it turns that anti-money-laundering legislation prevents it in at least some countries.
But BMWs and Lexuses will get you laid.
Are you saying that a titan won't? Damn.
It's an improvement on the old situation because events were timed in real time even with bad server lag, so if you have a 30-second cooldown on some ability or other, but a 5 minute lag, you only get to use it one tenth as many times as you're supposed to, which breaks the game design and is really infuriating to play. Now they slow the cooldowns down along with the lag in responding to commands so you can still keep doing stuff as much as the designers intended you to be able to. Kind of like the difference between the whole game slowing down and just dropping frames to compensate for lag in an FPS -- dropping frames is the easiest design solution, but if it happens too much it'll just get you killed. Slowing the entire game down lets you carry on playing even when the lag's really bad.
Hell, with the prodicts you mentioned you're not even ALLOWED to try to figure out how they work. That's called reverse engineering and it's against the license.
Here in the UK, the right to reverse engineer is legally mandated by statute, so they can't take it away with license terms.
and
are both parts of our statutes.
I was thinking along the same lines. Kudos and attaboys to Google for doing this, but what this gift unintentionally did was cause the schools to need to buy I/O devices for those machines. Why not 15000 OLPCs, which are 'complete'?
Because OLPCs have a different focus. OLPC is intended as a general purpose learning tool, while RPi is intended specifically to teach electronics & computer science. Presumably goolgle wanted the latter outcome rather than the former.
While I do want a Simcity game, as SimCity 4 is 9 years old at this point, I can't see this new product as a Simcity game. There is no local saves. [...] At this rate, I doubt it will have any fun cheat codes as was a staple of the series.
Cheat codes? Who needs 'em? I always used to edit the save file and give myself a virtually unlimited budget. Guess that one's gone, too...
google before you post OP. sweet fuck.
To be fair on OP, it seems that at least some of the applications they want to run are running under Windows and/or Mac environments, for which this solution does not apply.
Virtualizing a single application's windows from a remote machine is a non-trivial task that AFAIK hasn't been implemented in open source software for either of these platforms. The closest you'll get is by virtualizing the entire OS -- VirtualBox with Windows guests (and Windows only) can do this. You'd then have to run the virtualbox virtual machine process as an X client, and use X-over-SSH forwarding as described in many existing posts to get the windows to appear on the target machine. Performance will be poor (although my one experience of citrix suggests its performance was equally bad, so maybe you can tolerate that).
Read the friendly manual!
Given that the desktops are almost certainly running Windows, this by itself is only half of the solution. The other half is cygwin's X server with the -rootless or -multiwin command line options (the former requires you to run a native X11 window manager, the latter uses Windows' desktop window manager to provide window decorations).
Whilst individual islamic groups may do these things, they are not usually supported by the official churches of their religion. Most Muslims condemn such actions as not in accord with their actual religious beliefs. Scientology has made this an official policy, and abuse like this goes to the highest level. You'd find it hard to find actual practicing scientologists who would speak out against such acts.
Arduino from any other AVR board. It is a C-like language with a library; you're not programming to the bare board, you're not even writing your own main() routine. It is not intended for profressional programers, the target audience appears to be "multidisciplinary" (ie, people who aren't programmers).
I've been programming professionally for fifteen years - in C, Perl, PHP, Javascript, VB6, Actionscript, and other languages. Being a programmer, I was glad I didn't have to learn both embedded systems and assembler at the same time. For a guy like me, at a point where I've done just a little bit of kernel ptogramming for example, Arduino was really nice. "mov 0x40 0xD0" isn't what most programmers are familiar with.
I've been working directly with AVR chips rather than an Arduino on several projects over the last 6 months, and have never once had to touch an assembler. C is perfectly adequate for everything you're likely to want to do.
Raspberry Pi and Beaglebone both run Linux. The RPi is super cheap but is better targeted at apps which require a GUI. The BeagleBoard is more expensive but is better tuned for embedded use. It would be nice if the inverse were true, but oh well.
The Olimex A13-OLinuxIno is a reasonable compromise -- priced between the two boards, but much closer to the feature set of the beagleboard (and actually a faster processor with more memory than either). It was intended as a Linux-running competitor for Arduino, and is pretty nifty, as long as you're happy with the fact that it runs on a cheap chinese SoC processor that is only officially supported as a platform for making low-end Android tablets, so the Ubuntu port is not officially supported. And like Arduino, it's an open-hardware design. The only downside as far as I'm concerned is that it doesn't support either DVI or HDMI -- the only way of connecting a display is either via the integrated LCD panel port or on VGA, and resolution is limited to 800x600. But those constraints are fine for embedded use.
Erm...full disclosure, I worked in casinos, and also don't feel like being constantly under surveillance, either...
Just WHERE in a casino can you WORK and not be under constant surveillance?
In the surveillance room?
I wish to ${deity} that routers had a "reverse https proxy" function that would accept inbound https connections, strip the ssl, and transparently forward the traffic to the same port of an internal IP address where there's a device that's too stupid to know how to do SSL.
Have you considered setting up a VPN? Routers with integrated VPN functions are affordable these days (e.g. http://www.google.co.uk/products/catalog?q=dsl+router+vpn&sugexp=chrome,mod%3D11&um=1&ie=UTF-8&cid=11302817784067722053&sa=X&ei=Z3UHUfSWJrGp0AWNzYCwAw&ved=0CGMQ8wIwAw ). Alternatively, it wouldn't be too hard to set up the system you describe on a server inside your network and just forward your ports on the router to that system.
Bullshit. If your device has a reason to create an outbound connection, it is (for the most part) limited to one connection to one place for a specific purpose. (Disregarding intentionally buggered on-board software designed with malicious intent).
You're disregarding exactly the situation the GGP post was describing as the reason he turned UPNP off. GP's reply was a reasonable response: if you're assuming that software inside your network is malicious, it doesn't need UPNP to cause mischief... it'll probably hook up to an IRC server or similar in order to accept incoming commands, so that isn't a good reason to disable UPNP.
Now, this situation is (presumably) not malicious, but that doesn't make GP's response invalid. OTOH, I have to query how rare situations like this are. Very few devices automatically create firewall holes for themselves without user confirmation. Most UPNP routers make it very easy to monitor what holes you do have. The proportion of such devices that have massive security flaws like this is also likely to be low. I'm not therefore convinced that this situation is, on balance, enough to make me want to turn UPNP off.