Further, github has for all it's image, kept all it's stuff closed source. They chat up open source and say how great it is... Until it comes to the code they run on their servers, then they just don't say anything....
Contrast to gitlab which offers enough to let you make your own site if you prefer.
If it is to screw things over, look no further than the decline of SourceForge. At one point, their position of 'go-to place for open source projects' seemed unassailable. Then they died off and github became the new hotness in *very* short order. The kicker is that sourceforge technically gave a lot more services than github ever did, so projects were willing to give up having integrated hosting, powerful download management, and many other things. Also, a lot of projects were still using svn, so they had to go the extra mile to migrate to git.
Now look at github. By and large, projects use them for *a* git clone. Yes, the pull requests are useful in the context of the networking effect of the community, but generally speaking, a project could migrate to another similar service like gitlab or bitbucket without so much as even logging into their github account ever again. There is very very little 'stickiness' for github from a technical standpoint.
As far as Microsoft's track record for acquisitions, it's mixed. Skype clearly came out worse for the wear, unable to match competition and screwed up by MS' ambitions for it. On the other hand, LinkedIn still seems to be doing ok, and MS has seemingly not done too much to it yet.
I personally prefer gitlab anyway (I can actually self-host gitlab if I want to, unlike github), so I'm hoping this move makes gitlab more popular.
There are straightforward situations that are better resolved through exchange of text. When that happens, the conversation is searchable in the future.
For complex situations with a lot of back and forth, a phone call is warranted. Often alongside a text channel to copy paste things back and forth, or to send a picture or share screen. Always I like to lead this with a text conversation, so I know the context of the conversation going in, and to do things like "I'm in a meeting, I can't speak now, will reach out in 15 minutes".
In your example "Do you know about the issue at MZR?" is either a way to declare the topic of the conversation or to genuinely find out answer. For example your answer can be "yes, what do you need to know?" or "no, do you need help?" or "I know of it, but you'll want to speak with Fred, he's the one taking care of that". If he's looking for someone who already knows about it, then "no" is a helpful answer, or being redirected.
Particularly if your need is not urgent, then a text based message allows the user to respond when they can.
If I had to let all the conversations that are attempted with me over a lunch break go through voice mail, I'd spend a great deal of my time sorting through crap. Especially since a lot of people feel compelled to be extra verbose on the phone especially when leaving a message (because they really don't know what to do without back and forth conversation). Certainly they can similarly be verbose in text, but it's much less of an impact to deal with long text than a 3 minute voicemail.
My personal opinion is the same. My experience has been a generally glitchy platform.
It runs Android slower.
I don't know, but I would wonder if that is the case, is it because of how many ChromeOS systems are Intel versus ARM. The Android ecosystem is heavily ARM centric and Intel platforms can run it, but I just don't think they get the same attention.
It runs in landscape only,
At least our families lenovo chromebook does portrait
it has the fixed keyboard attached.
That is currently the choice of the hardware makers at the moment. Chromebooks that fold back (our family has a 'thinkpad' yoga chromebook, which has a crappy pointing device and no trackpoint, not very thinkpad-y) exist today to be tablet-like, though they don't commit to being totally keyboard free, which makes them pretty heavy compared to how it could be.
It has a windowing interface that is terrible to use with fingers on a touch device.
I concur, but I haven't seen a windowing interface that wasn't terrible to use on a touch device, and Android's windowing system is worse. Unless you mean to say a windowing system doesn't belong on such a form factor, in which case I understand and yes Android fits that sentiment well. I would agree with not having a windowing system prominent in a touch oriented device if that's your stance.
I will say I also don't get Google's obsession with ChromeOS. From a business perspective it doesn't do all that great. From a brand strength perspective, Android is the *much* stronger brand. In education ChromeOS has found something of a niche success, but they would probably do just as well or better transitioning that market to "new, Android with windowing support that works well on laptop form factor" platform with full compatibility for the google services that really drive the education market.
Of course, it does lock people into the Google story. Multiple hardware vendors is nice and all, but the software and services lock in is far more insidious than hardware lock in.
That's the shame of the market, you have to pick which of the three scary corporations you want to lock yourself into (google, microsoft, or apple). The only areas of computing where this isn't the case are large enterprises (that can roll their own stuff) and hobbyists/enthusiasts (that can use a rather untethered linux distro).
I don't trust Samsung or Amazon, and have no interest in being locked into their ecosystem and don't consider them trustworthy.
I'm surprised that you don't trust Samsung, but *do* trust* Google. Or maybe you are saying with respect to getting timely updates?
I personally am concerned about Samsung's market share, and I'm concerned that broadly, *no* other brand of Android device *consistently* has a track record of timely updates. Currently Motorola is occasionally with the program (e.g. the Moto X4 has had a pretty good track record so far, but that seems to be the exception rather than the rule)
I believe the data shows that the honeymoon is over for iPads as well as other devices.
I think the problem with tablets as a business market is that the use case is such that any old device can pretty much do what people want out of a tablet (read static content, play videos), there's no carrierd pushing for 'new free device every two years' and this means ultimately people aren't upgrading their tablets.
The tablet is an awkward middle. Too bulky and large to be on your person at all times, which means that novel sensors and applications aren't applicable to them. The size is also too big for comfortable game playing. On the flip side, it's smaller than a TV or a monitor and a good 'tablet' is touchscreen focused, and that's just too limited for content creation or higher end gaming.
It's a shame because particularly for reading comics and similar, it really is an awesome form factor. Windows tablets tend to be a tad big, but more critically, there are much better applications in android for this sort of media. Given that is pretty much the only use case that I go to my tablet for now, that's not a particularly exciting market to sell to.
This was true back in the day compared to vinyl, but there was a reason cassettes didn't totally replace vinyl records and technology has unambiguously done better than tapes now.
A-F) Presumably in contrast to vinyl G) Apart from streaming, even commercial audio files have no DRM, on ta technical level it's equally tradeable, on a legal level, those mix tapes were no better or worse. H) Though true, there are a lot of 'computer' devices, more than there ever were dual-deck casette player/recorders which were pretty ubiquitous. I) Apart from some Sony stuff, there really wasn't a widely used alternative that wasn't this. J) True of microsd cards K) Meh. Tape decks at best had noise reduction to try to fix the audio, "track selection" involved a great deal of time, and a tape deck would and did decide to destroy tapes or at least make you have to get out a pencil and start manually rolling the tape.
I suspect Vinyl resurgance is about being "quaint and cute" and about the rather generous real estate for nice album art, which is otherwise a pretty dead medium (yes, m4a files can embed album art, almost nothing shows it and when it does, it's a postage stamp sized rendering)..
Perhaps so, but I'll give him the benefit of the doubt that perhaps there's some complication since he did do a pretty impressive job. Of course he also says he could have done better on PPU fetch pattern prediction, which may be what you refer to.
Since it's just a fun stunt to pull, not like it has to be perfect anyway.
Well, as mentioned in the video, the glitchiness is because he's having to guess what the address will be before it actually gets read from, modifying his guess when predicted wrong for the next pass. Yes, he mentioned some facets of the experience due to the non-RT nature of the platform, but largely it's because of the having to guess addresses to respond to in advance.
In essence, the NES is a particularly convoluted video output converter here. A very impressive and difficult way to do something relatively simple, largely to make a point about imagining technology augmented brain.
Rental car companies demand even more stripped down models than the manufacturers are willing to offer for direct sales to cut costs. Hard to imagine them swallowing a $700 device to track a car when that is already a pretty saturated field with cheaper devices. Further, given the presumed interest in theft recovery, putting it in the license plate, the first part of a car ditched after being stolen, seems like a poor choice to integrate the tracking facility.
Well, I'd say that's only true in the context of Android, which isn't reassuring given the stuff on top of linux to worry about...
On the server side, Linux largely averted Windows drinking all the server milkshake in the first place (if the choices were commercial Unix on locked in platforms and Windows on inter operable hardware, Windows would have won). Windows did pretty much take over the groupware and directory roles, much to the chagirn of Novell. However, they have used their warchest to basically join in the cloud game so that even those pesky people running linux servers are now likely to be giving microsoft money to run their linux servers.
On the desktop side, while I hate to admit it, Linux desktops have not made an appreciable dent in the market, even when counting ChromeOS. OSX has about 10% which is something, but MS has 82% and there's no sign of any movement..
Of course you can replace "tech [industry]" with pretty much any industry...
This is the challenge with unlimited large business, when they take the lead they get so much in terms of resources, they can pretty much do whatever they want. If a promising project comes to disrupt their position, they can just toss a few billion dollars at it and take it over, to either use it or shut it down at their discretion.
Realistically speaking, an array of more numerous, yet smaller rotors may alleviate the noise and provide redundancy. This is one of the few new things that is easily done now that wasn't feasible years ago in aviation (computer assistance to translate human input into how to individually control the rotors).
Of course, 70 years back we had a lot of assumptions and seemingly viable concepts around 'flying cars', so there's a much longer tradition of optimism exceeding reality in the case of aviation.
I would say one key difference in practice is tolerance of poor maintenance.
In a car with poor maintenance and upkeep, you end up on the side of the road.
In an aircraft with poor maintenance, you die.
Of course this is not a common criteria, for flying cars people generally think about it being street legal and able to fly/land wherever. By that criteria, the helicopter can't do a grocery trip, at least not by itself, the last mile would have to be some other vehicle.
Guess the short of it is multi-motor, multi-rotor aircraft are going to be needed for the common slack user to be safer, and for it to tolerate moter/rotor failures and compensate.
The issue is that in practice this feature (and features like it in other languages) 99% of the time the programmer intends it to be used for persisting "boring old data" in the laziest way possible. The feature of having data be evaluated with executable instructions being honored is just a huge liability.
The 1% of the time when the programmer explicitly does use the "data can have code-to-eval" capability, it has been in my experience, always done better another way (such code is generally a pain to debug because an intuitive path is used for some of the code that executes, and I've yet to see a situation where they really *needed* to accomplish their goals that specific way.
Basically, it's encouraging a bad practice of mixing and matching data and executable code. Ideally you want your.jpg to be "just an image" and only have to worry about arbitrary executable data when dealing with an executable file.
Why would two robots talk on the phone in English, then they can talk REST?
REST is the web kids applying the long lived Unix philosophy of 'everything is a file' to web servers and pretending they are the first to think of it. Not a bad concept, but leaves a bad taste in my mouth everytime I hear it as it feels like it dismisses the folks who had that sorted out long age.
Also, why would two robots talk in REST, when that design is more for the humans, not for the software. Software is perfectly happy with much more efficient representations of data, so long as both ends agree upon a vocabulary up front, but it would have to do that for REST as well anyway.
Note that a lot of the developer community has made a pretty reasonable ask, for the browser to prompt the user and make it transparent to the web developer by default. Still auto-mute, but have a default UI in the browser to ask user to unmute the tab.
If course, a charging station at the most extreme is 15x slower than pumping the gas.
Which again would be fine if every restaurant I might want to stop at for a long haul has an appropriately potent charging station.
Charging will be a longer downtime, but because it's not as much a burden as vending gasoline, it is at least possible for restaurants to integrate the perk.
In my case, I did entertain offers and got much higher offers, but thus far my current employer when faced with that has always counter-offered with even more money.
So you can even stay in one place *and* get substantial increases, but you have to get the offers to induce the counter offers.
Of course, I don't think those 'old days' stable employment were ever marked by big pay increases either.
There's capacity and there's convenience/time to go a bit further. For you if you have a 400 mile trip to make, you bear the burden of a quick stop for gas.
For an electric car, even availing itself of tesla supercharger network, you will have to contend with a more inconvenient stop along the way.
Now this may be worth it in exchange for commuter experience of never having to stop for gas as you charge at home overnight, but we have to be honest that for long distance trips, ICE has both a capacity and refuel advantage still yet.
In the general case this may be true, but under 'check out this stunt' conditions, you can generally eliminate most of the factors that demand that.
I can promise you the tesla does not have as much traction as a 54 ton tractor, and neither did the porsche that did the same thing, or the vw, or the various strongmen that have made videos of pulling large planes.
I'm less concerned with commuting (I do have ways to charge at home, overnight, sure apartment dwellers may have a challenge, but homeowners are in decent shape here) and more the long haul trips.
Slashdot Mirror
Further, github has for all it's image, kept all it's stuff closed source. They chat up open source and say how great it is... Until it comes to the code they run on their servers, then they just don't say anything....
Contrast to gitlab which offers enough to let you make your own site if you prefer.
If it is to screw things over, look no further than the decline of SourceForge. At one point, their position of 'go-to place for open source projects' seemed unassailable. Then they died off and github became the new hotness in *very* short order. The kicker is that sourceforge technically gave a lot more services than github ever did, so projects were willing to give up having integrated hosting, powerful download management, and many other things. Also, a lot of projects were still using svn, so they had to go the extra mile to migrate to git.
Now look at github. By and large, projects use them for *a* git clone. Yes, the pull requests are useful in the context of the networking effect of the community, but generally speaking, a project could migrate to another similar service like gitlab or bitbucket without so much as even logging into their github account ever again. There is very very little 'stickiness' for github from a technical standpoint.
As far as Microsoft's track record for acquisitions, it's mixed. Skype clearly came out worse for the wear, unable to match competition and screwed up by MS' ambitions for it. On the other hand, LinkedIn still seems to be doing ok, and MS has seemingly not done too much to it yet.
I personally prefer gitlab anyway (I can actually self-host gitlab if I want to, unlike github), so I'm hoping this move makes gitlab more popular.
It really depends...
There are straightforward situations that are better resolved through exchange of text. When that happens, the conversation is searchable in the future.
For complex situations with a lot of back and forth, a phone call is warranted. Often alongside a text channel to copy paste things back and forth, or to send a picture or share screen. Always I like to lead this with a text conversation, so I know the context of the conversation going in, and to do things like "I'm in a meeting, I can't speak now, will reach out in 15 minutes".
In your example "Do you know about the issue at MZR?" is either a way to declare the topic of the conversation or to genuinely find out answer. For example your answer can be "yes, what do you need to know?" or "no, do you need help?" or "I know of it, but you'll want to speak with Fred, he's the one taking care of that". If he's looking for someone who already knows about it, then "no" is a helpful answer, or being redirected.
Particularly if your need is not urgent, then a text based message allows the user to respond when they can.
If I had to let all the conversations that are attempted with me over a lunch break go through voice mail, I'd spend a great deal of my time sorting through crap. Especially since a lot of people feel compelled to be extra verbose on the phone especially when leaving a message (because they really don't know what to do without back and forth conversation). Certainly they can similarly be verbose in text, but it's much less of an impact to deal with long text than a 3 minute voicemail.
I half agree with you.
It's shit.
My personal opinion is the same. My experience has been a generally glitchy platform.
It runs Android slower.
I don't know, but I would wonder if that is the case, is it because of how many ChromeOS systems are Intel versus ARM. The Android ecosystem is heavily ARM centric and Intel platforms can run it, but I just don't think they get the same attention.
It runs in landscape only,
At least our families lenovo chromebook does portrait
it has the fixed keyboard attached.
That is currently the choice of the hardware makers at the moment. Chromebooks that fold back (our family has a 'thinkpad' yoga chromebook, which has a crappy pointing device and no trackpoint, not very thinkpad-y) exist today to be tablet-like, though they don't commit to being totally keyboard free, which makes them pretty heavy compared to how it could be.
It has a windowing interface that is terrible to use with fingers on a touch device.
I concur, but I haven't seen a windowing interface that wasn't terrible to use on a touch device, and Android's windowing system is worse. Unless you mean to say a windowing system doesn't belong on such a form factor, in which case I understand and yes Android fits that sentiment well. I would agree with not having a windowing system prominent in a touch oriented device if that's your stance.
I will say I also don't get Google's obsession with ChromeOS. From a business perspective it doesn't do all that great. From a brand strength perspective, Android is the *much* stronger brand. In education ChromeOS has found something of a niche success, but they would probably do just as well or better transitioning that market to "new, Android with windowing support that works well on laptop form factor" platform with full compatibility for the google services that really drive the education market.
Yeah, things counting by units sold still show a decline, for everyone.
we all benefit
Of course, it does lock people into the Google story. Multiple hardware vendors is nice and all, but the software and services lock in is far more insidious than hardware lock in.
That's the shame of the market, you have to pick which of the three scary corporations you want to lock yourself into (google, microsoft, or apple). The only areas of computing where this isn't the case are large enterprises (that can roll their own stuff) and hobbyists/enthusiasts (that can use a rather untethered linux distro).
I don't trust Samsung or Amazon, and have no interest in being locked into their ecosystem and don't consider them trustworthy.
I'm surprised that you don't trust Samsung, but *do* trust* Google. Or maybe you are saying with respect to getting timely updates?
I personally am concerned about Samsung's market share, and I'm concerned that broadly, *no* other brand of Android device *consistently* has a track record of timely updates. Currently Motorola is occasionally with the program (e.g. the Moto X4 has had a pretty good track record so far, but that seems to be the exception rather than the rule)
I believe the data shows that the honeymoon is over for iPads as well as other devices.
I think the problem with tablets as a business market is that the use case is such that any old device can pretty much do what people want out of a tablet (read static content, play videos), there's no carrierd pushing for 'new free device every two years' and this means ultimately people aren't upgrading their tablets.
The tablet is an awkward middle. Too bulky and large to be on your person at all times, which means that novel sensors and applications aren't applicable to them. The size is also too big for comfortable game playing. On the flip side, it's smaller than a TV or a monitor and a good 'tablet' is touchscreen focused, and that's just too limited for content creation or higher end gaming.
It's a shame because particularly for reading comics and similar, it really is an awesome form factor. Windows tablets tend to be a tad big, but more critically, there are much better applications in android for this sort of media. Given that is pretty much the only use case that I go to my tablet for now, that's not a particularly exciting market to sell to.
This was true back in the day compared to vinyl, but there was a reason cassettes didn't totally replace vinyl records and technology has unambiguously done better than tapes now.
A-F) Presumably in contrast to vinyl
G) Apart from streaming, even commercial audio files have no DRM, on ta technical level it's equally tradeable, on a legal level, those mix tapes were no better or worse.
H) Though true, there are a lot of 'computer' devices, more than there ever were dual-deck casette player/recorders which were pretty ubiquitous.
I) Apart from some Sony stuff, there really wasn't a widely used alternative that wasn't this.
J) True of microsd cards
K) Meh. Tape decks at best had noise reduction to try to fix the audio, "track selection" involved a great deal of time, and a tape deck would and did decide to destroy tapes or at least make you have to get out a pencil and start manually rolling the tape.
I suspect Vinyl resurgance is about being "quaint and cute" and about the rather generous real estate for nice album art, which is otherwise a pretty dead medium (yes, m4a files can embed album art, almost nothing shows it and when it does, it's a postage stamp sized rendering)..
Perhaps so, but I'll give him the benefit of the doubt that perhaps there's some complication since he did do a pretty impressive job. Of course he also says he could have done better on PPU fetch pattern prediction, which may be what you refer to.
Since it's just a fun stunt to pull, not like it has to be perfect anyway.
Well, as mentioned in the video, the glitchiness is because he's having to guess what the address will be before it actually gets read from, modifying his guess when predicted wrong for the next pass. Yes, he mentioned some facets of the experience due to the non-RT nature of the platform, but largely it's because of the having to guess addresses to respond to in advance.
In essence, the NES is a particularly convoluted video output converter here. A very impressive and difficult way to do something relatively simple, largely to make a point about imagining technology augmented brain.
Rental car companies demand even more stripped down models than the manufacturers are willing to offer for direct sales to cut costs. Hard to imagine them swallowing a $700 device to track a car when that is already a pretty saturated field with cheaper devices. Further, given the presumed interest in theft recovery, putting it in the license plate, the first part of a car ditched after being stolen, seems like a poor choice to integrate the tracking facility.
Well, I'd say that's only true in the context of Android, which isn't reassuring given the stuff on top of linux to worry about...
On the server side, Linux largely averted Windows drinking all the server milkshake in the first place (if the choices were commercial Unix on locked in platforms and Windows on inter operable hardware, Windows would have won). Windows did pretty much take over the groupware and directory roles, much to the chagirn of Novell. However, they have used their warchest to basically join in the cloud game so that even those pesky people running linux servers are now likely to be giving microsoft money to run their linux servers.
On the desktop side, while I hate to admit it, Linux desktops have not made an appreciable dent in the market, even when counting ChromeOS. OSX has about 10% which is something, but MS has 82% and there's no sign of any movement..
Of course you can replace "tech [industry]" with pretty much any industry...
This is the challenge with unlimited large business, when they take the lead they get so much in terms of resources, they can pretty much do whatever they want. If a promising project comes to disrupt their position, they can just toss a few billion dollars at it and take it over, to either use it or shut it down at their discretion.
Solution: dirigible. Nice and quite.
Realistically speaking, an array of more numerous, yet smaller rotors may alleviate the noise and provide redundancy. This is one of the few new things that is easily done now that wasn't feasible years ago in aviation (computer assistance to translate human input into how to individually control the rotors).
Of course, 70 years back we had a lot of assumptions and seemingly viable concepts around 'flying cars', so there's a much longer tradition of optimism exceeding reality in the case of aviation.
I would say one key difference in practice is tolerance of poor maintenance.
In a car with poor maintenance and upkeep, you end up on the side of the road.
In an aircraft with poor maintenance, you die.
Of course this is not a common criteria, for flying cars people generally think about it being street legal and able to fly/land wherever. By that criteria, the helicopter can't do a grocery trip, at least not by itself, the last mile would have to be some other vehicle.
Guess the short of it is multi-motor, multi-rotor aircraft are going to be needed for the common slack user to be safer, and for it to tolerate moter/rotor failures and compensate.
The issue is that in practice this feature (and features like it in other languages) 99% of the time the programmer intends it to be used for persisting "boring old data" in the laziest way possible. The feature of having data be evaluated with executable instructions being honored is just a huge liability.
The 1% of the time when the programmer explicitly does use the "data can have code-to-eval" capability, it has been in my experience, always done better another way (such code is generally a pain to debug because an intuitive path is used for some of the code that executes, and I've yet to see a situation where they really *needed* to accomplish their goals that specific way.
Basically, it's encouraging a bad practice of mixing and matching data and executable code. Ideally you want your .jpg to be "just an image" and only have to worry about arbitrary executable data when dealing with an executable file.
Why would two robots talk on the phone in English, then they can talk REST?
REST is the web kids applying the long lived Unix philosophy of 'everything is a file' to web servers and pretending they are the first to think of it. Not a bad concept, but leaves a bad taste in my mouth everytime I hear it as it feels like it dismisses the folks who had that sorted out long age.
Also, why would two robots talk in REST, when that design is more for the humans, not for the software. Software is perfectly happy with much more efficient representations of data, so long as both ends agree upon a vocabulary up front, but it would have to do that for REST as well anyway.
Note that a lot of the developer community has made a pretty reasonable ask, for the browser to prompt the user and make it transparent to the web developer by default. Still auto-mute, but have a default UI in the browser to ask user to unmute the tab.
If course, a charging station at the most extreme is 15x slower than pumping the gas.
Which again would be fine if every restaurant I might want to stop at for a long haul has an appropriately potent charging station.
Charging will be a longer downtime, but because it's not as much a burden as vending gasoline, it is at least possible for restaurants to integrate the perk.
In my case, I did entertain offers and got much higher offers, but thus far my current employer when faced with that has always counter-offered with even more money.
So you can even stay in one place *and* get substantial increases, but you have to get the offers to induce the counter offers.
Of course, I don't think those 'old days' stable employment were ever marked by big pay increases either.
There's capacity and there's convenience/time to go a bit further. For you if you have a 400 mile trip to make, you bear the burden of a quick stop for gas.
For an electric car, even availing itself of tesla supercharger network, you will have to contend with a more inconvenient stop along the way.
Now this may be worth it in exchange for commuter experience of never having to stop for gas as you charge at home overnight, but we have to be honest that for long distance trips, ICE has both a capacity and refuel advantage still yet.
In the general case this may be true, but under 'check out this stunt' conditions, you can generally eliminate most of the factors that demand that.
I can promise you the tesla does not have as much traction as a 54 ton tractor, and neither did the porsche that did the same thing, or the vw, or the various strongmen that have made videos of pulling large planes.
I'm less concerned with commuting (I do have ways to charge at home, overnight, sure apartment dwellers may have a challenge, but homeowners are in decent shape here) and more the long haul trips.