And all of that existed with upstart, daemontools, many worked with inetd, and in general many other systems solved the same problem. Systemd's danger is its insistence on doing _everything_, including replacing logging, DHCP, NTP, and numerous other already working and well understood components. Its proponents have made clear that they intend to replace all of/etc.
Discussing systemd as if it were only an init script replacement is missing out on why people are upset about it.
If the version presented to the Congress member has substantial changes written in, just for that Congressman, or if the treaty is modified without notification, then what the Congressman thought they were agreeing to will not match what the treaty. That is _begging_ for abuse, much like a recent project I saw where the code compiled by the developer bore only a passing resemblance to what was in the source control and which had been planned for release.
Police can't normally walk away from the scene, and they are compelled to attend in the first place. That does not mean I support this law, but police injury rates are _much_ higher than most work. According to http://www.governing.com/gov-d..., they're only surpassed by nursing care, and I can easily believe that.
> If it wasn't for Microsoft, we would still be on mainframes and mini-computers. Paying jacked up prices. For crap, frankly.
Nonsense. Apple demonstrated, at a similar time, that personal computing was affordable. Bill Gates and Microsoft successfully assembled the business suite that helped drive PC sales, but there were other designs and even operating systems evolving that were compete.
Also note, Microsoft did not "create" the 64-bit core of Windows NT and Windows XP. They lifted a great deal of it, wholesale, from VMS with the help of Dave Cutler when they hired him and his development away from DEC. The lawsuits over this were fascinating, but please give credit where credit is due.
Facebook could help schools far, far more by enforcing their minimum age requirements of 13. I'm seeing far younger kids sucked into their computers by the Facebook chat, and refusing to go outside or explore knowledge outside their own little clique of online "likes".
Goodness, you're reaching back. But I also remember Trumpet Winsock quite well, and also the superior TCP stacks from FTP Software that preceded it and the collapse of that company when Microsoft released their own, inferior TCP stack. If we're going to compare the length of computing experience to prove whose memory is correct, I remember quite well the awkwardness of installing getting DEC hardware support for systems that were running BSD releaes, the sales of the first Macintosh personal computers, and the fascination when Tim Berners Lee first published HTML and the "world wide web" was born.
Those were heady days, and very exciting for nerds and geeks who were tasked to get it all working together. I can reach back further, but no one was paying me before those experiences. But I suspect your experience of the browser wars does not predate mine.
With that in mind, I was referring to specific periods of browser evolution, not to the earliest history. In particular, I was referring to the tendency of OEMs to provide multiple bulky subtly incompatible browser versions on the same hardware at sales time. Chrome is only the lastest among the suite of such tools to be included by OEMs. And the only one that has been _mandated_ for desktops has been Internet Explorer on Windows systems. Chrome, at least, can be gracefully removed. Removing IE is not graceful, by deliberate design. It was a key point in the Microsoft monopoly lawsuits.
Its understandable to be bothered by the bloat of a browser you didn't want. But compared to the IE desktop monopoly abuses, Google and Chrome have been quite polite.
I'm afraid your "history" is too recent. You seem to be referring to the feature filled but UI altering rewrites over the last 5 years. I was referring to the even older history, where Microsoft _lost_ various lawsuits about their abuse of their monopoly to enforce the use of Internet Explorer in the USA, back in 2001.
If you'd please stop cursing, you can look at the court history and the hands-on memory of admins, including me. The fraud by Microsoft about inability to remove IE was a problem. The punitive licensing for OEM's that dared to include Netscape on their systems were profound, non-technological abuses by Microsoft against Netscape. The requirement of IE to access Microsoft updates was a third.factor, partly technological, but primarily a policy decision. Netscape market share grew in spite of these illegally monopolistic practices, so they clearly used to have serious advantages, and your claims that older versions of IE were ever "not nearly as buggy" are not based on any reviews or experience I can find. It worked well with Microsoft's own web server, but both the web server and web client violated published standards at every opportunity.
The "works best in IE" coding practice is documented in software I continue to work with, 20 years after it was written, and was in place when it was first released. It was a problem for web authors who actually followed the RFC's and various coding practices.
You seem to remember the Netscape vs. IE vs. Opera wars rather differently than I do. I admit that Netscape was quite buggy until it went open source with Netscape 4, when it improved and stabilized rapidly.
It's also often a corporate standard, especially for companies and their clients with older, Windows specific software tools. And many proxies are configured to lie about the web client they are proxying for, in order to provide access to upstream websites which demand IE. There are many examples, such as:
My concern was more that AT&T is unusually cooperative with federal monitoring of communications. There are reasons to want the data, and even good uses for it such as letting this senior AOL user has accidentally built up an extraordinary phone bill. But there are so many _bad_ reasons to want the data, such as personal or meta analysis of anyone with political or media use of personal telephones, that I'm forced to remember that AT&T has cooperated in aggressive secret domestic surveillance programs with no apparent objections.
I've no personal ability to prevent the analysis or meta-analysis of customer data: I'm concerned with possible, even likely, abuse of the information.
You have valid logical and legal points. But the acts would be considered felonies if committed by an adult. The subtle difference between a delinquency and a felony in the early stages are subtle enough not to confuse the typical reader with.
Why would the EFF be involved? They're not "the protectors of all technology", there has to be someone whose rights they'd be defending. That would be the owner of the open source, or possibly free software, copyrights. And given that the first case the EFF ever helped litigate was against the US Secret Service, and that the lawyer involved is still there, I think they'd be willing to take on Goldman Sachs if needed.
Unfortunately, none of the copyrights or of open source or 'copylefts' of "free as in speech" software require returning your modifications to the world at large. You _only_ have to provide them, if asked, to the client you wrote them for or provide binaries for, and that requirement applies only to "free software", not to most "open source". The open source licenses generally _deliberately_ allow you to proprietize your personal version of the software, and do not require you to publish changes, precisely so that the "business model" of providing your own secret enhancements can continue. This is the core of a great many open source projects funding, including Citrix Xen, Zmanda, and Sun Microsystems before Oracle bought them.
This one way relationship with free software is extremely common, especially in the financial software community. I and my colleagues are considered fairly odd because we publish our patches, and send them upstream to the code authors. We sometimes have real difficulty negotiating IP clauses in contracts with others because they do not understand, or profoundly fear, the practice. And the difficulty is not with the engineers doing the work. It's with their own lawyers, frightened of providing any information which could be used as any kind of leverage at all in any other case, and their managers, who argue that if they are not _compelled_ to, why should they provide this assistance to competitors?
My team tries to sell it on the basis of supportability. If they patches and improvements get published upstream, other eyes can enhance it, they don't have to maintain and keep spending resources to maintain their own in-house fork, and upsream changes are much less likely to become incompatible with their in-house changes. Then, if necessary, we point out examples, ideally with software they're already using. But it wastes a lot of early setup time getting this negotiated.
I can think of half a dozen companies I worked with in the last decade, all of whom thought they'd re-invented network protocols. All of them found that by the time they'd implemented necessary error correction, buffering, and re-transmit protocols for missed data that they'd actually _lost_ performance. It never showed up in the early testing because the inexperienced, "key developer" didn't know the history or the available technologies, so they'd never tested it under realistic circumstances.
Having their billing system trigger a flag when it hits 10x the usual cost and halt access and red flag for support to call them when it hits 100x is NOT hard or invasive.
"Halting access" would mean cutting off their landline service. That is _not_ something to do lightly to someone in the midst of their personal or business crisis.
Many people in the USA live in remote areas with no good cell coverage and no cable or fiber optic. And that $50/month undoubtedly includes unrestricted local calls, which is why his previous always-on AOL usage was not increasing his bill.
There used to be good commercial reasons for AT&T to monitor for this sort of thing. It was tying up long distance trunk lines nearly 24x7 for a month, and those used to be radically more expensive and less available than local connections. It required an actual physical copper connection from one part of the country to the other, maintained 24x7 with significant electrical and maintenance costs. But today, with all the different ISP's and phone services handling Voice Over IP instead and making much larger data connections, a single landline phone connection is a few packets lost in the flood of data.
Also, monitoring for this kind of accident is paying a lot more attention to individual customer bills and usage than I necessarily want AT&T monitoring. AT&T has already established that they cooperate extensively with monitoring US communications at NSA request, especially with the notorious "Room 641A". DO we want them collecting and acting on this kind of data?
One of my and my group's major sources of income is cleaning up after those "one developer" projects. The "rone developer" often has no idea how, or no willingness, to set up a testing plan before releases, to integrate robust security, to make software high availability, or to scale it behind a certain very modest size.
The result is that the first project or demo works well and is very lean and agile in the performance sense. But as the number of customers grow, or as people find and report bugs, scaling up and keeping it working well is much easier for the larger, more cautious team. Ideally, they code reviewed each other's work and pointed out where a fix here broke a feature elsewhere, or pointed out the edge cases that also need to be handled. As an example, what works on a laptop sitting next to the server running the multi-player game may not work so well behind three firewalls, NAT, and an overburdened local cable network setup. Lone developers often are not expected to spend time on those issues.
And the lawyers, who billed for the time filing the case. And the clients who, I suspect, worked with the lawyers to find the friendliest venue to file the case ina: a friendly venue can make an enormous difference in court cases. It's called "forum shopping", and it's a critical tactical factor.
I'm afraid, son or daughter, that this is why some of us oldsters started learning the software that runs various robots about 25 years ago. I'm also afraid that your bet on that "geezer" card is making me raise the pot by about 10 years and call. Unless someone else wants to bid in this game?
The key when confronted by awareness of changing technology and workplace requirements is not to bemoan who will have the power. It's to join, or at least make sure you have the skills to work for, the powerful.
I'll offer a list of components I dearly missed when last using a shop.
0.01" permanent markers, very useful for drawing fine lines to cut or marking places to drill.
Thread gauge, because people will drop bolts and screws and get confused about which bin they go in.
Hot knife for cutting plastics, especially if it's hut enough to cut Teflon.
Velcro ty-wraps, especially those that come in the cheap big pre-perforated rolls.
Furniture clamps, for gluing bulky items.
Ziploc bags for keeping components together when people leave the workshop.
Voltmeter, one that can safely measure 120 Volt, and fuses for it when unskilled people misuse it.
Good pair of diagonal wire cutters.
Dremel tool.
Bins to put different projects in, rather than leaving them out on the bench.
Examine carefully the 'Trusted Computing' hardware and software components for new computers. Governmental agencies already have access to not only the escrowed keys, but to the master keys used to revoke and authorize other new keys. For personal security, it's quite troubling.
Slashdot Mirror
And all of that existed with upstart, daemontools, many worked with inetd, and in general many other systems solved the same problem. Systemd's danger is its insistence on doing _everything_, including replacing logging, DHCP, NTP, and numerous other already working and well understood components. Its proponents have made clear that they intend to replace all of /etc.
Discussing systemd as if it were only an init script replacement is missing out on why people are upset about it.
If the version presented to the Congress member has substantial changes written in, just for that Congressman, or if the treaty is modified without notification, then what the Congressman thought they were agreeing to will not match what the treaty. That is _begging_ for abuse, much like a recent project I saw where the code compiled by the developer bore only a passing resemblance to what was in the source control and which had been planned for release.
Police can't normally walk away from the scene, and they are compelled to attend in the first place. That does not mean I support this law, but police injury rates are _much_ higher than most work. According to http://www.governing.com/gov-d..., they're only surpassed by nursing care, and I can easily believe that.
> If it wasn't for Microsoft, we would still be on mainframes and mini-computers. Paying jacked up prices. For crap, frankly.
Nonsense. Apple demonstrated, at a similar time, that personal computing was affordable. Bill Gates and Microsoft successfully assembled the business suite that helped drive PC sales, but there were other designs and even operating systems evolving that were compete.
Also note, Microsoft did not "create" the 64-bit core of Windows NT and Windows XP. They lifted a great deal of it, wholesale, from VMS with the help of Dave Cutler when they hired him and his development away from DEC. The lawsuits over this were fascinating, but please give credit where credit is due.
Facebook could help schools far, far more by enforcing their minimum age requirements of 13. I'm seeing far younger kids sucked into their computers by the Facebook chat, and refusing to go outside or explore knowledge outside their own little clique of online "likes".
Goodness, you're reaching back. But I also remember Trumpet Winsock quite well, and also the superior TCP stacks from FTP Software that preceded it and the collapse of that company when Microsoft released their own, inferior TCP stack. If we're going to compare the length of computing experience to prove whose memory is correct, I remember quite well the awkwardness of installing getting DEC hardware support for systems that were running BSD releaes, the sales of the first Macintosh personal computers, and the fascination when Tim Berners Lee first published HTML and the "world wide web" was born.
Those were heady days, and very exciting for nerds and geeks who were tasked to get it all working together. I can reach back further, but no one was paying me before those experiences. But I suspect your experience of the browser wars does not predate mine.
With that in mind, I was referring to specific periods of browser evolution, not to the earliest history. In particular, I was referring to the tendency of OEMs to provide multiple bulky subtly incompatible browser versions on the same hardware at sales time. Chrome is only the lastest among the suite of such tools to be included by OEMs. And the only one that has been _mandated_ for desktops has been Internet Explorer on Windows systems. Chrome, at least, can be gracefully removed. Removing IE is not graceful, by deliberate design. It was a key point in the Microsoft monopoly lawsuits.
Its understandable to be bothered by the bloat of a browser you didn't want. But compared to the IE desktop monopoly abuses, Google and Chrome have been quite polite.
We complained, believe me. That's why I was so fond of my old self-winding wristwatch. It wasn't as precise as some, but but it was quite effective.
I'm afraid your "history" is too recent. You seem to be referring to the feature filled but UI altering rewrites over the last 5 years. I was referring to the even older history, where Microsoft _lost_ various lawsuits about their abuse of their monopoly to enforce the use of Internet Explorer in the USA, back in 2001.
If you'd please stop cursing, you can look at the court history and the hands-on memory of admins, including me. The fraud by Microsoft about inability to remove IE was a problem. The punitive licensing for OEM's that dared to include Netscape on their systems were profound, non-technological abuses by Microsoft against Netscape. The requirement of IE to access Microsoft updates was a third.factor, partly technological, but primarily a policy decision. Netscape market share grew in spite of these illegally monopolistic practices, so they clearly used to have serious advantages, and your claims that older versions of IE were ever "not nearly as buggy" are not based on any reviews or experience I can find. It worked well with Microsoft's own web server, but both the web server and web client violated published standards at every opportunity.
The "works best in IE" coding practice is documented in software I continue to work with, 20 years after it was written, and was in place when it was first released. It was a problem for web authors who actually followed the RFC's and various coding practices.
You seem to remember the Netscape vs. IE vs. Opera wars rather differently than I do. I admit that Netscape was quite buggy until it went open source with Netscape 4, when it improved and stabilized rapidly.
According to William Seabrook, who had more range of diet than many human meat eaters, it is pretty good and tastes like high quality veal.
http://www.smithsonianmag.com/...
It's also often a corporate standard, especially for companies and their clients with older, Windows specific software tools. And many proxies are configured to lie about the web client they are proxying for, in order to provide access to upstream websites which demand IE. There are many examples, such as:
http://unix.stackexchange.com/...
My concern was more that AT&T is unusually cooperative with federal monitoring of communications. There are reasons to want the data, and even good uses for it such as letting this senior AOL user has accidentally built up an extraordinary phone bill. But there are so many _bad_ reasons to want the data, such as personal or meta analysis of anyone with political or media use of personal telephones, that I'm forced to remember that AT&T has cooperated in aggressive secret domestic surveillance programs with no apparent objections.
I've no personal ability to prevent the analysis or meta-analysis of customer data: I'm concerned with possible, even likely, abuse of the information.
Unlike all the others? The most infamous such case was that of Microsoft and Internet Explorer.
http://en.wikipedia.org/wiki/U....
You have valid logical and legal points. But the acts would be considered felonies if committed by an adult. The subtle difference between a delinquency and a felony in the early stages are subtle enough not to confuse the typical reader with.
Why would the EFF be involved? They're not "the protectors of all technology", there has to be someone whose rights they'd be defending. That would be the owner of the open source, or possibly free software, copyrights. And given that the first case the EFF ever helped litigate was against the US Secret Service, and that the lawyer involved is still there, I think they'd be willing to take on Goldman Sachs if needed.
Unfortunately, none of the copyrights or of open source or 'copylefts' of "free as in speech" software require returning your modifications to the world at large. You _only_ have to provide them, if asked, to the client you wrote them for or provide binaries for, and that requirement applies only to "free software", not to most "open source". The open source licenses generally _deliberately_ allow you to proprietize your personal version of the software, and do not require you to publish changes, precisely so that the "business model" of providing your own secret enhancements can continue. This is the core of a great many open source projects funding, including Citrix Xen, Zmanda, and Sun Microsystems before Oracle bought them.
This one way relationship with free software is extremely common, especially in the financial software community. I and my colleagues are considered fairly odd because we publish our patches, and send them upstream to the code authors. We sometimes have real difficulty negotiating IP clauses in contracts with others because they do not understand, or profoundly fear, the practice. And the difficulty is not with the engineers doing the work. It's with their own lawyers, frightened of providing any information which could be used as any kind of leverage at all in any other case, and their managers, who argue that if they are not _compelled_ to, why should they provide this assistance to competitors?
My team tries to sell it on the basis of supportability. If they patches and improvements get published upstream, other eyes can enhance it, they don't have to maintain and keep spending resources to maintain their own in-house fork, and upsream changes are much less likely to become incompatible with their in-house changes. Then, if necessary, we point out examples, ideally with software they're already using. But it wastes a lot of early setup time getting this negotiated.
I can think of half a dozen companies I worked with in the last decade, all of whom thought they'd re-invented network protocols. All of them found that by the time they'd implemented necessary error correction, buffering, and re-transmit protocols for missed data that they'd actually _lost_ performance. It never showed up in the early testing because the inexperienced, "key developer" didn't know the history or the available technologies, so they'd never tested it under realistic circumstances.
Having their billing system trigger a flag when it hits 10x the usual cost and halt access and red flag for support to call them when it hits 100x is NOT hard or invasive.
"Halting access" would mean cutting off their landline service. That is _not_ something to do lightly to someone in the midst of their personal or business crisis.
Many people in the USA live in remote areas with no good cell coverage and no cable or fiber optic. And that $50/month undoubtedly includes unrestricted local calls, which is why his previous always-on AOL usage was not increasing his bill.
There used to be good commercial reasons for AT&T to monitor for this sort of thing. It was tying up long distance trunk lines nearly 24x7 for a month, and those used to be radically more expensive and less available than local connections. It required an actual physical copper connection from one part of the country to the other, maintained 24x7 with significant electrical and maintenance costs. But today, with all the different ISP's and phone services handling Voice Over IP instead and making much larger data connections, a single landline phone connection is a few packets lost in the flood of data.
Also, monitoring for this kind of accident is paying a lot more attention to individual customer bills and usage than I necessarily want AT&T monitoring. AT&T has already established that they cooperate extensively with monitoring US communications at NSA request, especially with the notorious "Room 641A". DO we want them collecting and acting on this kind of data?
One of my and my group's major sources of income is cleaning up after those "one developer" projects. The "rone developer" often has no idea how, or no willingness, to set up a testing plan before releases, to integrate robust security, to make software high availability, or to scale it behind a certain very modest size.
The result is that the first project or demo works well and is very lean and agile in the performance sense. But as the number of customers grow, or as people find and report bugs, scaling up and keeping it working well is much easier for the larger, more cautious team. Ideally, they code reviewed each other's work and pointed out where a fix here broke a feature elsewhere, or pointed out the edge cases that also need to be handled. As an example, what works on a laptop sitting next to the server running the multi-player game may not work so well behind three firewalls, NAT, and an overburdened local cable network setup. Lone developers often are not expected to spend time on those issues.
And the lawyers, who billed for the time filing the case. And the clients who, I suspect, worked with the lawyers to find the friendliest venue to file the case ina: a friendly venue can make an enormous difference in court cases. It's called "forum shopping", and it's a critical tactical factor.
When it's connected to an implanted insulin pump, it's controlling lives pretty directly:
http://www.washingtonpost.com/...
The classic video of how employers can commit H1B fraud is at:
https://www.youtube.com/watch?...
What they describe is how to skirt the law, but still hire the less expensive H1B that an employer wants. to quote:
"Our goal, clearly, is not to find a qualified and interested US worker."
I'm afraid, son or daughter, that this is why some of us oldsters started learning the software that runs various robots about 25 years ago. I'm also afraid that your bet on that "geezer" card is making me raise the pot by about 10 years and call. Unless someone else wants to bid in this game?
The key when confronted by awareness of changing technology and workplace requirements is not to bemoan who will have the power. It's to join, or at least make sure you have the skills to work for, the powerful.
Real programmers use:
https://xkcd.com/378/
I'll offer a list of components I dearly missed when last using a shop.
0.01" permanent markers, very useful for drawing fine lines to cut or marking places to drill.
Thread gauge, because people will drop bolts and screws and get confused about which bin they go in.
Hot knife for cutting plastics, especially if it's hut enough to cut Teflon.
Velcro ty-wraps, especially those that come in the cheap big pre-perforated rolls.
Furniture clamps, for gluing bulky items.
Ziploc bags for keeping components together when people leave the workshop.
Voltmeter, one that can safely measure 120 Volt, and fuses for it when unskilled people misuse it.
Good pair of diagonal wire cutters.
Dremel tool.
Bins to put different projects in, rather than leaving them out on the bench.
Examine carefully the 'Trusted Computing' hardware and software components for new computers. Governmental agencies already have access to not only the escrowed keys, but to the master keys used to revoke and authorize other new keys. For personal security, it's quite troubling.