I posted a similar comment to this effect. Also a PhD holder, and if I answer a question directly and without pause, it means I really know a good answer. Otherwise, I will ask some questions and explore like everybody else.
Seriously, I've had people tell me recursion is terrible and knew nothing about tail call optimization, or if they heard of that, they don't understand exactly what it means. Never mind I took advanced coursework on programming language semantics where we had to formally define it. Now, if you asked me for a formal definition, I would look that up, because it's been a while,
Add in the fact I don't have the PhD from the right schools, so for a few, it can't be that useful, and I really wish it wasn't near impossible to find academic appointments.
I have an advanced degree in Computer Science, yet for various reasons, I'm in the market. I usually just have to bite my tongue and go through the paces, but I've never been in the room with anybody that understand algorithms to the degree that I do that asked these questions. Because the people that have the same understanding that do look at my education and just skip to what I'm interested in, what they are looking for and so on.
Once, I decided to use recursion just for a change of pace. In the end, the interviewer just showed that he didn't actually know what tail call form was. Remember, the people that really know what they are talking about will always admit what they don't know and what they look up. It's the one that are trying to be and look clever that are problematic.
Also, if you are just trying to see how people approach problems, just say so. Also, the vast majority of programming isn't oriented around algorithms and data structures, it's all domain representation and library integration. But nobody asks "let's try to capture the essential properties and methods of a class for a shopping site customer." for example.
Neurons aren't digital processors. A set of connected neurons isn't either. Neuroscience already knows that it's really difficult to learn about the structure and function of the brain from the available tools. What was more interesting was that they were able to pick up anything. They found that the chip had a master clock, for example.
There are people already challenging the use of viewing the brains as a computer (signal ins and outs) in terms of really understanding how brains organize and function. So, given all this, it's not surprising that the methods didn't fare well. The neuroscientists already knew they had a very tough task, it's those in CS and AI that are assuming that understanding the brain is the same as understanding a collection of digital circuits.
Type checking is ultimately formal verification (see the Curry-Howard correspondence). Of course, this limits what type systems can specify about a program, and practically, the restrictions placed on languages based on certain more advanced type systems greatly limit their use in general programming.
The modern version is much more of a reference than textbook. While exercises still are part the book, it really for testing knowledge. Also, the field has just exploded and the task is really daunting. Satisfiability is an example, the current fascicle is 320 pages, but a more in depth look at the problem could run to twice that much and more. And there is new work being done on the topic every day.
Frankly, less and less programmers will need it, as it is easier and easier to create and use algorithmic libraries. Also, optimization just isn't what it used to be. Modern embedded silicon is quite roomy and complete control of the hardware leads to more problems (think buffer overflows) than it often solves. Finally, code that is verifiable is of more overall value, and so simpler algorithms and more constricted languages may be the future of a lot of code bases.
While I agree that they can be difficult, they exist for a reason, and frankly, with some proper study of functional and parallel programming, they can be used correctly. Closures have been a part of programming since LISP and frankly, higher order functions and functional programming can solve a lot of problems quite nicely. Certainly, mutability makes that a bit more difficult, but it can be managed. Finally, figuring out what variables are captured in a closure is quite doable by static analysis, which would be a nice addition to some programming environments.
Threads are more difficult for sure, but there are frameworks that can provide some useful higher level abstractions to make there use a bit easier. Lower level usage can be quite difficult, but a bit of actual thought and design goes along way. Also, there are some really powerful tools for multithreaded debugging, they just aren't freely available.
Mainly, I've found that people that can't manage these abstractions just don't have enough background in operating systems (which also discusses basics of concurrency) and functional programming, topics that used to be required in CS education. They can be learned and it can increase your overall code quality and reduce some frustrations.
It's about the increasing biases in the industry that assumes that older programmers just can't possibly pick up new technology without a lot of help. It's quite the opposite in many cases. As if somebody that started programming hasn't moved from language to language multiple times. They understand the fundaments, and they don't just chase one trend after another. They have a good sense of what is mature enough for consideration and what isn't. They know that programming all the time not only isn't necessary, it is detrimental in the long term.
Seriously. The author seems to have no idea what the Windows Subsystem for Linux is nor how it works. There are separate file systems for a reason. The things the author tries are exactly the things Microsoft make it very clear will be very problematic at best.
The POSIX model and the Windows model of permissions are completely different. Instead of trying to map between the two (something Cygwin does variably well, sometimes really badly), they provide a file system with the semantics needed. Finally, there is no Microsoft version of Bash. It's Bash, compiled for the Linux kernel (like most packages) that is being run as is by the subsystem. The subsystem has bugs, this is known, this is beta software and it is software that even Microsoft is smart enough to know that they need as many people trying it to find bugs as needed.
And it does fill a very specific need and more importantly, it's further building out the newer subsystem parts of NT, which might come in handy for other things down the road.
Also, you don't have the right to deny your child the chance to be healthy and not suffer preventable diseases for any reason. We are a modern society. This is simple, just take the choice away. A child that medically able to be vaccinated will be, free of charge. Any attempt to interfere with that process is child neglect and will be handled as such.
Also, I want vaccine availability to be a specific line item in foreign aid for countries. I want a consistent vaccine development program at the CDC with proper funding.
Why all this? Because vaccines are one of the most successful public health interventions ever. Only basic sanitation and clean water has been more successful and yes, they are in the same category.
On what service?. Azure has CDN services with similar pricing structure as AWS. No free tier, though.
Of course, cloud pricing is very tricky, overall, but the cost structures between Amazon and Azure are more and more in line these days, The competition between the two is starting to show some pricing benefits.
And is limited to how well you use it. As the article noted, OpenOffice and LibreOffice will do they same thing as well. They noted Sheets doesn't, but I can't get Google Sheets to handle dates consistently at all sometimes.
This is just an excellent example of what works for a large population of users can be a bad thing for a small set of users. After this paper, I expect the error rate to drop dramatically, given how easy the fix is.
Okay, they cancel Broxton, but then they release this. So, smartphones and tablets are out, but this is a great prototyping board for industrial IoT and other smart devices? Look, if they don't have a story on cellular network capabilities, nobody is going to care, and if they do have a story there, then they didn't really leave those markets. Does the Surface Phone crawl along, zombie like, after all? At any rate, Intel has a lot of work to do in the embedded space. A lot.
So, it'll run on Windows 10, including mobile, and I expect anybody that's on an older Windows Phone will be on a newer Windows 10 phone or something else by then.
Already use it, have for years, but the number of the ports varies and often lag behind in versioning. Certain tools work better than others overall. Not Cygwin's fault, they have to try and emulate certain semantics via Win32 in user space, and some calls, permissions, etc. just don't work quite right. (fork is an excellent example). But, this is because it is exceptionally hard to do. I am extremely grateful to Corinna Vinschen and her team for the tireless work she has done for countless years to bring a usable Unix like command line to Windows. She's was tirelss in tracking NT kernel semantics and the Win32 user space. But, a proper kernel subsystem brings it to another level and that is the better overall approach. And this time, Microsoft is working with good partners to get things working. But, I expect Cygwin (and MSYS) to have a role for quite some time to come.
It's still marked as a beta right now. Hope they push hard and get into general availability this year. It's useful. Running unmodified console mode apps from the Ubuntu user space is a useful thing.
Right now, if secured boot is off, this policy doesn't kick in. That may change of course. For the vast majority of Windows users, this is fine, but for power users, kind of a pain.
The basics of journalism do help. Just dumping raw data with no concern to how it may affect third parties that are irrelevant to the main story really hurts your overall credibility. Not even showing any attempt to verify the information as valid (because it is easy to tamper with digital information) with additional sources does as well. News matters. Providing a context to a given set of information is important. Asking for comment and/or rebuttal from various parties is important, even if they refuse. Showing judgement as to what is relevant is important.. Not doing so opens them up to a ton of valid criticism. Some editorial prudence would go a long way overall.
I think it's a much needed expansion of the.Net ecosystem (better late than never) and I do think will become a useful alternative to the JVM, which Oracle seems to have little interest in evolving or improving. It took forever to get invokedynamic added as an opcode. Tail call optimization is still not supported, after years of being requested. And there's tons of other ideas on the table that aren't getting anywhere.
In the case of.Net core, it's all open source. The runtime, the compiler, the cli tools. Sure, Microsoft isn't going to take any proposal on the table, but there's a process for making changes. And, C# is a great language to develop in (and F# is nice when you need it). And who knows, maybe it'll be a Scala target some day. I honestly think people will be surprised at it's performance compared to the JVM. It's adapted a lot of modernization that the JVM eschews for backwards compatibility and known predictability.
I'm sure there will be appeals and Oracle will keep pushing the issue, but for now, it's a good thing. Frankly, it'll keep people using Java for longer, which helps Oracle a bit. Of course, Oracle and IBM are just fine locking people into insanely expensive middleware platforms too.
Personally. I'd more than like to see Java fade into maturity. It's clunky, verbose and many of the frameworks they use are showing their age. Of course, I'm biased having done recent work in C#. I've just gotten very used to it's asynchronous programming features and frameworks that embrace that and higher order programming. And.Net core is removing a major annoyance by finally going more OS neutral.
I do hope this will prompt Google to make Go a first class option for Android. Swift worked out well for Apple. I know Java has it's fan, but the ones I talk still act like Sun is in charge of the platform. They aren't and it really shows. Oracle has been a terrible steward of Java and the Java platform. Even IBM was better, but not by much. I'm surprised that more Java programmers aren't frustrated by how things have been managed, but imagine most of them aren't aware of how competing languages are changing.
I can use OS/X, Linux. With all the fervor over Windows 10, there's still Windows options to reduce or turn off telemetry off (in some versions). Google's been doing this forever, making billions for it, and there's no escaping it. Why won't Microsoft get in on the trend to make a better OS?
No option to self host your own Google software, no way to get them to truly honor your preference not to track you, nothing. I can't even pay them to do so. And if my employer or school uses their applications, I have to trust them that they don't track those users, but if some of the current lawsuits against them turn out to be true, that trust was misplaced.
Look, if you want to make software services, just do so. But Google can't let go of ads or advertising revenue and are dragging other software companies with them. Frustrating. But, go ahead, keep using Chrome and making fun of MS or Apple for having their own browsers and cheer as their market share goes down.
It'd be nice if Microsoft took a stand on this, but why when everybody else is doing it and too few people will even notice (the Slashdot crowd isn't enough) Look at the number of pre-installed applications on most mobile phones. All the software installed by OEMs on top of Windows. Apple moving more and more to the App Store for OS/X. The vast majority of these new devices have ad-driven software on them, but everybody is just used to it. A few more pre installed apps won't even get noticed.
As for businesses, it seems pretty clear that MS wants to move people towards Enterprise just for the chance to get away from most of this stuff. Personally, I've given up on this not getting worse. Google this huge wave of applications that were "free" as long as you agreed to ads, to demographic information gathering, and now, all the major players want in. I can only hope that when the free upgrade window ends, more users start yelling and if we are lucky, Windows 10 Professional turns back into what it should be, a OS with all the controls to opt out and just let me do my job present.
But, the advantage Apple has always had is a very small set of hardware configurations, but if you let a bug like this out the door, you aren't taking nearly as much advantage of that as you should. So, actually, I do hold Apple to a higher standard here, because they need to be. You don't get to set exactly what computers run your OS, and how much they will be and then turn around and say, "oh, that obvious bug we let out the door, it happens, what can you do"
Take the ASUS motherboard UEFI boot problem. Microsoft had admit the problem, post a workaround, and didn't even mention the thousands and thousands of other hardware combinations that worked just fine, because nobody cares anyway. But, if I was an OSX users, I would be upset. The damn things should just work and trudge along, update after update for a few years. But they increasing don't do that, and people paid the premiums for the platform anyway and thank Apple for it. Just stop thanking them for a start.
Sure, as soon as we get a peer reviewed set of studies that make that conclusion. By the way, this study doesn't do that at all. "Every genetic factor we found is so small that known environmental factors like socioeconomic status on the same measures would completely overwhelm the genetic variation, making any actual genetic contribution almost unnoticeable" isn't "people are just born smarter"
The infrastructure around Git is just too large and has too many large adopters. And it's not like there aren't open source alternatives (Mercurial) that people would shift in the unlikely case that Git imploded. Also, I doubt any remember, but SourceGear tried to make it's own open source DVCS (darcs) that is still going on too.
But, you never can tell who will be at the top of the hill. I remember when ClearCase was considered to be the top notch VCS, and it did have some very nice features in the day. But, things change.
The only game changer I can see is tools that know what it is tracking has structure beyond files. For example, it'd be nice to track changes to the level of a block in a function in a class in a project, not just file changes. And tools that allow for queries on that structure (show me every class method you can find with more than six parameters on it, for example). But, code repositories are hard to do in the general case.
Slashdot Mirror
I posted a similar comment to this effect. Also a PhD holder, and if I answer a question directly and without pause, it means I really know a good answer. Otherwise, I will ask some questions and explore like everybody else.
Seriously, I've had people tell me recursion is terrible and knew nothing about tail call optimization, or if they heard of that, they don't understand exactly what it means. Never mind I took advanced coursework on programming language semantics where we had to formally define it. Now, if you asked me for a formal definition, I would look that up, because it's been a while,
Add in the fact I don't have the PhD from the right schools, so for a few, it can't be that useful, and I really wish it wasn't near impossible to find academic appointments.
I have an advanced degree in Computer Science, yet for various reasons, I'm in the market. I usually just have to bite my tongue and go through the paces, but I've never been in the room with anybody that understand algorithms to the degree that I do that asked these questions. Because the people that have the same understanding that do look at my education and just skip to what I'm interested in, what they are looking for and so on.
Once, I decided to use recursion just for a change of pace. In the end, the interviewer just showed that he didn't actually know what tail call form was. Remember, the people that really know what they are talking about will always admit what they don't know and what they look up. It's the one that are trying to be and look clever that are problematic.
Also, if you are just trying to see how people approach problems, just say so. Also, the vast majority of programming isn't oriented around algorithms and data structures, it's all domain representation and library integration. But nobody asks "let's try to capture the essential properties and methods of a class for a shopping site customer." for example.
Neurons aren't digital processors. A set of connected neurons isn't either. Neuroscience already knows that it's really difficult to learn about the structure and function of the brain from the available tools. What was more interesting was that they were able to pick up anything. They found that the chip had a master clock, for example.
There are people already challenging the use of viewing the brains as a computer (signal ins and outs) in terms of really understanding how brains organize and function. So, given all this, it's not surprising that the methods didn't fare well. The neuroscientists already knew they had a very tough task, it's those in CS and AI that are assuming that understanding the brain is the same as understanding a collection of digital circuits.
Type checking is ultimately formal verification (see the Curry-Howard correspondence). Of course, this limits what type systems can specify about a program, and practically, the restrictions placed on languages based on certain more advanced type systems greatly limit their use in general programming.
The modern version is much more of a reference than textbook. While exercises still are part the book, it really for testing knowledge. Also, the field has just exploded and the task is really daunting. Satisfiability is an example, the current fascicle is 320 pages, but a more in depth look at the problem could run to twice that much and more. And there is new work being done on the topic every day.
Frankly, less and less programmers will need it, as it is easier and easier to create and use algorithmic libraries. Also, optimization just isn't what it used to be. Modern embedded silicon is quite roomy and complete control of the hardware leads to more problems (think buffer overflows) than it often solves. Finally, code that is verifiable is of more overall value, and so simpler algorithms and more constricted languages may be the future of a lot of code bases.
But, it's an awesome project and always will be.
While I agree that they can be difficult, they exist for a reason, and frankly, with some proper study of functional and parallel programming, they can be used correctly. Closures have been a part of programming since LISP and frankly, higher order functions and functional programming can solve a lot of problems quite nicely. Certainly, mutability makes that a bit more difficult, but it can be managed. Finally, figuring out what variables are captured in a closure is quite doable by static analysis, which would be a nice addition to some programming environments.
Threads are more difficult for sure, but there are frameworks that can provide some useful higher level abstractions to make there use a bit easier. Lower level usage can be quite difficult, but a bit of actual thought and design goes along way. Also, there are some really powerful tools for multithreaded debugging, they just aren't freely available.
Mainly, I've found that people that can't manage these abstractions just don't have enough background in operating systems (which also discusses basics of concurrency) and functional programming, topics that used to be required in CS education. They can be learned and it can increase your overall code quality and reduce some frustrations.
It's about the increasing biases in the industry that assumes that older programmers just can't possibly pick up new technology without a lot of help. It's quite the opposite in many cases. As if somebody that started programming hasn't moved from language to language multiple times. They understand the fundaments, and they don't just chase one trend after another. They have a good sense of what is mature enough for consideration and what isn't.
They know that programming all the time not only isn't necessary, it is detrimental in the long term.
Seriously. The author seems to have no idea what the Windows Subsystem for Linux is nor how it works. There are separate file systems for a reason. The things the author tries are exactly the things Microsoft make it very clear will be very problematic at best.
The POSIX model and the Windows model of permissions are completely different. Instead of trying to map between the two (something Cygwin does variably well, sometimes really badly), they provide a file system with the semantics needed. Finally, there is no Microsoft version of Bash. It's Bash, compiled for the Linux kernel (like most packages) that is being run as is by the subsystem. The subsystem has bugs, this is known, this is beta software and it is software that even Microsoft is smart enough to know that they need as many people trying it to find bugs as needed.
And it does fill a very specific need and more importantly, it's further building out the newer subsystem parts of NT, which might come in handy for other things down the road.
Also, you don't have the right to deny your child the chance to be healthy and not suffer preventable diseases for any reason. We are a modern society. This is simple, just take the choice away. A child that medically able to be vaccinated will be, free of charge. Any attempt to interfere with that process is child neglect and will be handled as such.
Also, I want vaccine availability to be a specific line item in foreign aid for countries. I want a consistent vaccine development program at the CDC with proper funding.
Why all this? Because vaccines are one of the most successful public health interventions ever. Only basic sanitation and clean water has been more successful and yes, they are in the same category.
On what service?. Azure has CDN services with similar pricing structure as AWS. No free tier, though.
Of course, cloud pricing is very tricky, overall, but the cost structures between Amazon and Azure are more and more in line these days, The competition between the two is starting to show some pricing benefits.
And is limited to how well you use it. As the article noted, OpenOffice and LibreOffice will do they same thing as well. They noted Sheets doesn't, but I can't get Google Sheets to handle dates consistently at all sometimes.
This is just an excellent example of what works for a large population of users can be a bad thing for a small set of users. After this paper, I expect the error rate to drop dramatically, given how easy the fix is.
Okay, they cancel Broxton, but then they release this. So, smartphones and tablets are out, but this is a great prototyping board for industrial IoT and other smart devices? Look, if they don't have a story on cellular network capabilities, nobody is going to care, and if they do have a story there, then they didn't really leave those markets. Does the Surface Phone crawl along, zombie like, after all? At any rate, Intel has a lot of work to do in the embedded space. A lot.
So, it'll run on Windows 10, including mobile, and I expect anybody that's on an older Windows Phone will be on a newer Windows 10 phone or something else by then.
Already use it, have for years, but the number of the ports varies and often lag behind in versioning. Certain tools work better than others overall. Not Cygwin's fault, they have to try and emulate certain semantics via Win32 in user space, and some calls, permissions, etc. just don't work quite right. (fork is an excellent example). But, this is because it is exceptionally hard to do. I am extremely grateful to Corinna Vinschen and her team for the tireless work she has done for countless years to bring a usable Unix like command line to Windows. She's was tirelss in tracking NT kernel semantics and the Win32 user space. But, a proper kernel subsystem brings it to another level and that is the better overall approach. And this time, Microsoft is working with good partners to get things working. But, I expect Cygwin (and MSYS) to have a role for quite some time to come.
It's still marked as a beta right now. Hope they push hard and get into general availability this year. It's useful. Running unmodified console mode apps from the Ubuntu user space is a useful thing.
Right now, if secured boot is off, this policy doesn't kick in. That may change of course. For the vast majority of Windows users, this is fine, but for power users, kind of a pain.
The basics of journalism do help. Just dumping raw data with no concern to how it may affect third parties that are irrelevant to the main story really hurts your overall credibility. Not even showing any attempt to verify the information as valid (because it is easy to tamper with digital information) with additional sources does as well. News matters. Providing a context to a given set of information is important. Asking for comment and/or rebuttal from various parties is important, even if they refuse. Showing judgement as to what is relevant is important.. Not doing so opens them up to a ton of valid criticism. Some editorial prudence would go a long way overall.
I think it's a much needed expansion of the .Net ecosystem (better late than never) and I do think will become a useful alternative to the JVM, which Oracle seems to have little interest in evolving or improving. It took forever to get invokedynamic added as an opcode. Tail call optimization is still not supported, after years of being requested. And there's tons of other ideas on the table that aren't getting anywhere.
In the case of .Net core, it's all open source. The runtime, the compiler, the cli tools. Sure, Microsoft isn't going to take any proposal on the table, but there's a process for making changes. And, C# is a great language to develop in (and F# is nice when you need it). And who knows, maybe it'll be a Scala target some day. I honestly think people will be surprised at it's performance compared to the JVM. It's adapted a lot of modernization that the JVM eschews for backwards compatibility and known predictability.
I'm sure there will be appeals and Oracle will keep pushing the issue, but for now, it's a good thing. Frankly, it'll keep people using Java for longer, which helps Oracle a bit. Of course, Oracle and IBM are just fine locking people into insanely expensive middleware platforms too.
Personally. I'd more than like to see Java fade into maturity. It's clunky, verbose and many of the frameworks they use are showing their age. Of course, I'm biased having done recent work in C#. I've just gotten very used to it's asynchronous programming features and frameworks that embrace that and higher order programming. And .Net core is removing a major annoyance by finally going more OS neutral.
I do hope this will prompt Google to make Go a first class option for Android. Swift worked out well for Apple. I know Java has it's fan, but the ones I talk still act like Sun is in charge of the platform. They aren't and it really shows. Oracle has been a terrible steward of Java and the Java platform. Even IBM was better, but not by much. I'm surprised that more Java programmers aren't frustrated by how things have been managed, but imagine most of them aren't aware of how competing languages are changing.
I can use OS/X, Linux. With all the fervor over Windows 10, there's still Windows options to reduce or turn off telemetry off (in some versions). Google's been doing this forever, making billions for it, and there's no escaping it. Why won't Microsoft get in on the trend to make a better OS?
No option to self host your own Google software, no way to get them to truly honor your preference not to track you, nothing. I can't even pay them to do so. And if my employer or school uses their applications, I have to trust them that they don't track those users, but if some of the current lawsuits against them turn out to be true, that trust was misplaced.
Look, if you want to make software services, just do so. But Google can't let go of ads or advertising revenue and are dragging other software companies with them. Frustrating. But, go ahead, keep using Chrome and making fun of MS or Apple for having their own browsers and cheer as their market share goes down.
It'd be nice if Microsoft took a stand on this, but why when everybody else is doing it and too few people will even notice (the Slashdot crowd isn't enough) Look at the number of pre-installed applications on most mobile phones. All the software installed by OEMs on top of Windows. Apple moving more and more to the App Store for OS/X. The vast majority of these new devices have ad-driven software on them, but everybody is just used to it. A few more pre installed apps won't even get noticed.
As for businesses, it seems pretty clear that MS wants to move people towards Enterprise just for the chance to get away from most of this stuff. Personally, I've given up on this not getting worse. Google this huge wave of applications that were "free" as long as you agreed to ads, to demographic information gathering, and now, all the major players want in. I can only hope that when the free upgrade window ends, more users start yelling and if we are lucky, Windows 10 Professional turns back into what it should be, a OS with all the controls to opt out and just let me do my job present.
Apologies in advance for the poor editing on my part before posting.
But, the advantage Apple has always had is a very small set of hardware configurations, but if you let a bug like this out the door, you aren't taking nearly as much advantage of that as you should. So, actually, I do hold Apple to a higher standard here, because they need to be. You don't get to set exactly what computers run your OS, and how much they will be and then turn around and say, "oh, that obvious bug we let out the door, it happens, what can you do"
Take the ASUS motherboard UEFI boot problem. Microsoft had admit the problem, post a workaround, and didn't even mention the thousands and thousands of other hardware combinations that worked just fine, because nobody cares anyway. But, if I was an OSX users, I would be upset. The damn things should just work and trudge along, update after update for a few years. But they increasing don't do that, and people paid the premiums for the platform anyway and thank Apple for it. Just stop thanking them for a start.
Sure, as soon as we get a peer reviewed set of studies that make that conclusion. By the way, this study doesn't do that at all. "Every genetic factor we found is so small that known environmental factors like socioeconomic status on the same measures would completely overwhelm the genetic variation, making any actual genetic contribution almost unnoticeable" isn't "people are just born smarter"
The infrastructure around Git is just too large and has too many large adopters. And it's not like there aren't open source alternatives (Mercurial) that people would shift in the unlikely case that Git imploded. Also, I doubt any remember, but SourceGear tried to make it's own open source DVCS (darcs) that is still going on too.
But, you never can tell who will be at the top of the hill. I remember when ClearCase was considered to be the top notch VCS, and it did have some very nice features in the day. But, things change.
The only game changer I can see is tools that know what it is tracking has structure beyond files. For example, it'd be nice to track changes to the level of a block in a function in a class in a project, not just file changes. And tools that allow for queries on that structure (show me every class method you can find with more than six parameters on it, for example). But, code repositories are hard to do in the general case.