The old unix way of separating things into different directories that are usually on different partitions or drives, has some advantages. Separating state files for the whole machine, state files per user, config files, and binaries (executable and libraries) that are shared between machines on NFS, is a big plus. OTOH, there is definite simplicity and manageability in having everything in one directory for a given application, except for common data more than one application might share.
What would be nice is a way to structure data in a tree such that it appears like like an application oriented structure, where each application has its branch, then underneath that are the different aspects, like binaries here, libraries there, data files, state files, config files, and so on, while still being able to put everything in a specific filesystem where it is manageable in certain ways. One possibility, though messy, is to have an application area under each aspect, like/usr/bin/${appname}/ or/usr/lib/${appname}/. This is often done, especially under/opt/. But Linux does have a way to organize it so you can see it both ways: bind mounts.
What would be needed for security is to make sure that only authorized locations (e.g. the corporate application repository) are accessed. I would never be able to suggest to management to use a system like this unless and until I can thoroughly vet this system for security. And I would never start that process until you can at least tell me you believe it does that. It might be cool from a geek perspective to be able to dynamically download on demand an application from anywhere on the net the author places it, but in business where the geeks have managed to get management to be security aware, such a thing just won't fly. Not only would management want to make sure users won't be running anything someone else might change on the fly, they also want to make sure users are not running anything not approved by management (this in a corporate office setting). If all this can be done restricted to a set of servers on which only approved and vetted applications reside, then you might have something here that is better than the current NFS mounting of / and/or/usr. But if you are designing this for geek coolness factor, it could likely not be suitable for business.
So what if I am installing some application which expects its files to be in various places like/usr/share or/etc/foo or wherever? Or is this concept only limited to applications that are aware of ROX?
Competing in this case is fruitless. You can't compete because at the wages that would be paid to do so, you can't afford anything in the US at all. Of course, you could live in luxury in India. Or wait 20 years until the US economy is dragged down to the level of India.
The real problem is not that India has competitors. The problem is that the exchange rate of the US Dollar is such that converting $10000 to Rupees gives someone there plenty enough money to live on very comfortably for a year. It's not a level playing field.
In time, it will level out as more Indians are employed to the point of exhausting its limited supply. India has a few very good schools, comparable to the few top schools in the US, but below that is not much else. In 10 years or so, IIT grads will be demanding higher salaries. I'm already seeing a wee bit of this now from recruiters in India finding it harder to find people to fill job slots.
The problem is, over this time frame, the US gets hurt because there will be far fewer people experienced (because they are not getting experience now) in tech areas, and fare fewer people graduating (because they are turning away from tech majors) with tech degrees. The US companies have always been hurting themselves by insisting on people who are already trained, educated, and/or experienced, and never willing to offer any of that themselves.
Still, the level of outsourcing to India is a small portion of the total unemployment and underemployment in the US. Corporations simply don't have any incentive to do anything other than what making profits gives them, and the US government is failing to provide incentives (like lowering taxes) for specific actions like hiring Americans. The real problem is not that India is able to compete; the real problem is that a US government is too stupid to create the kind of incentives for American businesses to hire more, and hire more Americans.
The benefits can work both ways, either way. With incentives to hire Americans here, the economy prospers because more consumers can buy more products. Business grows, and as the US unemployment diminishes, more employment overseas has to happen, either through US companies going back overseas (scaling back up what they do have), and more imports into the US from overseas. All the arguments about improving the economy work both ways. The issue for the US government is whether to let our own high tech skill base be ravaged in the process.
I build the mission critical programs from source code, and just let the rest be installed as binary packages. I build from source even if I don't need to just to be sure I won't have extra unexpected issues should I ever need to actually make modifications to source and rebuild. I really don't have very many local modifications, but I'm prepared just in case.
Additionally, I do this all on one master machine (with a backup of it kept live on another machine), build binary packages of my own from my source builds, and install those packages on the actual servers. That way I have even more consistency, though at the cost of ultimate optimization. But I think it is better to be able to quickly reinstall a machine, as well as use checksum verifications that there are no trojans.
I use Slackware, but this could be done with most systems, including FreeBSD, Linux (most distributions, including Debian and the RPM based ones), NetBSD, OpenBSD, and even Solaris.
Re:CSS does NOT always degrade gracefully with HTM
on
CSS for the LDP?
·
· Score: 1
However, using tables for layout is still technically compliant with the standards that specify the table tags. But it's just HTML compliant. The catch with CSS is that the level of "gracefully" degrading isn't satisfactory to lots of people, especially people who come from graphical media like print or video where they have always had explicit control. The only way to overcome this is to do a whole lot better job in two areas. The first is to promote CSS a lot better and more strongly include the requirement to not do the things in HTML that break sites that use CSS. And the other area is to promote better browsers that support CSS correctly, and are not obese. Just because you might have 2 gigabytes of RAM to run Mozilla or whatever does not mean everyone else does. I still have to use Netscape 4 because of these limitations, and some of my friends have even smaller computers than I do. CSS is broken in Netscape 4 and has to be left turned off. I do use Firebird occaisionally, but I can't use it regularly due to the bloat and quite many other bugs (including a form submission security bug). Maybe some day someone with real systems understanding will decide to put together a truly smart modular browser that won't bloat itself beyond what the current web sites it's viewing will need. I can give specific advice, but I won't except directly to the lead developers (because it is a bit complex and would require some back and forth discussion to make sure it is conveyed correctly).
Re:CSS does NOT always degrade gracefully with HTM
on
CSS for the LDP?
·
· Score: 1
The technology, or its documentation, or its promotion, is at fault when lots of people use it wrong.
If the technology is hard to use then the technology is at fault. For example, a lot of the problems I see are where specific table cells have a specific background color. I don't know that CSS can even specify that, and so maybe it has to be done in HTML, which in turn forces font colors to be done in HTML, too.
Absolutely none of the CSS promoting sites say anything about how to specifically deal with this problem, or with any other specific problem. They describe CSS, tell people it should be used, list the benefits, then leave people hanging with little more than some reference documentation and examples on a few pages that are relevant to real world problems.
Fix that... e.g. make documentation better... and maybe things will work out. Now all I need is a decent, reliable, and non-obese browser, that does handle CSS correctly (Netscape 4 doesn't fit the latter requirement, and Mozilla Firebird doesn't fit the former requirement).
The gentoo page spills for me, too. The TLDP page does not. That can't possibly be the same. Just because it might render the same for you doesn't mean it has to for everyone. But if it renders differently even for just a few, then there is at least something different that TLDP did better than gentoo.
Some of the Gentoo page is UNREADABLE
on
CSS for the LDP?
·
· Score: 1
Some of the Gentoo page is UNREADABLE. That is because it is attempting (well, succeeding) to change the background color in a table cell without changing the foreground color at the same time. The headings in the tables is where this problem is. Th background is dark blue and the foreground is black. What kind of silliness is that? This is NOT correct graceful degrading; it's screwed up. Of course the fix is easy... put the necessary font color setting in there. So why are they not doing it? Or do they just not care about older browsers and the people that have damned good reasons to use them.
So in the end, TLDP is more readable to many people.
CSS does NOT always degrade gracefully with HTML
on
CSS for the LDP?
·
· Score: 1
CSS does NOT always degrade gracefully with HTML... depending on how well the web developers handle it. If absolutely NO HTML presentation is specified at all, then the site does degrade. Whether that can be called graceful or not is a matter of opinion.
But once a site starts using even the slightest bit of presentation specification in HTML, then things get bad very quickly. For example if a background color is specified in a table cell, it might be in conflict with the default color used for say hyperlinks by the browser. So if the background color is specified, so must every affected font color.
Here is an example of a site, and how it looks in an older browser, where the developers claim they are using "web standards" to make a site, and it degrades horribly. Actually, I would call this site absolute crap. When I communicated with them about it, they simply claimed that it "meets federal accessibility standards". Technically, they are not actually doing web standards correctly. But the point is, there are too many webmasters who are idiots and can never get it right, so I have my doubts about promoting this concept.
Q: Why are the chinese communists so afraid of free exchange of ideas and criticism? A: They're afraid they'll have to give up power and find real jobs.
Oh, you mean like being paid $1.40 a day (14 hour shift) making US flags?
Here is my idea for a better model for Cable TV. If Congress is going to mandate things, this is what they should mandate.
Make Cable TV simply be a high speed internet where every program provider can make themselves available without having to contract with every individual Cable TV company. The program providers then choose whether to make their programming available free and unencrypted, or encrypted with different payment schemes such as pay per program or pay per month.
The whole idea is to separate the infrastructure provider from the content provider (or as in the case of electricity and gas, the energy provider). With the exception of requirements to carry the basic market over-the air TV stations, Cable TV service should be entirely ala-carte. With digital transmission and encryption like we have now, and computers to interface directly with users and control the access to programming, it should be a very cheap and easy thing to do in the long run.
Is this why EV1 hosts spammers? To make money? Oh yeah, there is money to be made in spam. And all you have to do is just ignore the complaints from other networks that lose money because of the abuse and attacks. Spammers eventually do leave because the IP addresses get majorly blocked, then EV1 can announce that a spammer was terminated an looks good because of that. Of course not everyone falls for it, so many networks have EV1 still blocked. I do, at least until Robert Marsh comes clean about spam.
As for his current situation with SCO... I'm not considering that to be a "agrees with the anti-SCO crowd" until he puts another million into the fund to defend other companies from SCO that now are under greater risk due to his stupid actions.
Robert Marsh has a lot of expensive cleaning up to do.
And apparently he has the business of hosting spammers, too.
Sign up a spammer and let them spam.
Ignore the complaints from other networks being attacked and losing money.
Make money.
Spammer gets tired of being blocked and moves on.
EV1 announces to world that they've terminated a spammer.
Another best of both worlds scenario. Except that because my network was one of those attacked, I blocked every EV1 address space and those blocks remain until EV1 becomes a clean operating hosting company.
Maybe some of the churn is from being spam-blocked? EV1 has a history of hosting spammers. My network was attacked by one of his spammers for 5 weeks continuously and they would never do a damned thing about it (and Robert Marsh would not take the phone call, either). So they stay blocked until they fix the problem (of not dealing with spam complaints properly), and Robert Marsh has to fix this personally, now.
But his network is NOT rock solid. It has a history of hosting spammers, and at least SMTP is block from that address space on many networks, including mine. The day Robert Marsh announces he will terminate all spammer customers, and accept no more spammers, is the day I look at removing that block. So if he can turn about on the SCO issue, maybe there's hope he can turn about on the spam issue. But his actions will speak more than his words.
Microsoft Office has no threat of viral infection. That's because viral infection is very real. Hell, they ought to remove all doubt and just ship Microsoft Office pre-infected.
I see no reason to even bother using Java anymore. It's been a royal pain where PHP and Python have not been. Open sourcing it may have breathed some life back into it, but... well... that didn't happen as I see.
College education really don't help you that much in the knowledge department. It should prepare you for life long on the job learning so you can adapt to all changes (and there will be constant changes). Sadly, while it is the case businesses do want people with genuine on the job experience, they are also unwilling to be the providers of that. Most of them want people that got their on the job experience somewhere else. They seem to also want people that won't learn anything new so they can just shove them out the door in a few years and get someone new and cheaper.
The sad fact of the matter is that CS grads are not qualified for most positions and won't get more than a glance by most recruiters.
Most positions require experience anyway. Recent grads don't have that. But if you're talking about positions specifically for non-experienced people, then show me those requirements so I can see if they are legitimate or not. I've seen way to many job postings that have unrealistically picky requirements, both of experienced and non-experienced people.
Keep in mind that for every person who does present a level of experience in a skill area, someone did have to hire them once without requiring that skill. They obviously acquired it somewhere (should be obvious where by their resume). Over the past several years, however, businesses have been less willing to hire someone who has to learn some of the needed skills on the job... and that means the pool of those people who have learned those skills on the job is shrinking.
... get an MBA
That's fine for some people. Hell it would be nice if more business executives really did understand the technology better, which would happen if more technical people went the MBA route. But the MBA path isn't right for everyone. Some people just love the technology too much to ever want to do the kind of work an MBA prepares them for. All you really need of these people is that they have a basic understanding of how their technical contribution makes business work. But this does require managers to respect them for their focus on technology (something they could not do as much if they are getting an MBA, too). These are not CTO/CIO jobs I'm talking about here.
Very few people get to write software from scratch nowadays.
That's one of the reasons I left software and went into network and security administration. I happen to love writing software, which is why I don't want to do that at work (it would ruin it for me).
I'll be much happier working 45 hours a week at 300 bucks an hour as a financial planner/estate planner while coding on the side than working 45 hours a week for someone else to maintain their code.
Exactly. Sadly, working for corporations these days tends to rule out doing a job you can love. It's just a job for money. And that is one of the reasons fewer people are going into high tech these days... because it isn't a career path for money anymore.
Just pray that they don't start outsourcing financial/estate planning to India, as they are already doing to investment analysis and tax preparation. But I'd say your new career path is safe for at least the next 4 or 5 years. Maybe.
You still have to use a valid HTTP "Host:" header with the correct hostname for the site to come up (that's how virtual by name works). And "www.mp3" doesn't work.
There might be a.ogg one, too. But I haven't found any roots for it.
The thing is, these don't need to be in some public root (ICANN, YOUCANN, or FOOCANN); they are intentionally hidden. And if you did stumble across the IP, you'd still have to know what "Host:" header domain name to use to get the site.
Who gives ICANN the right to dictate what TLDs will be usable for our DNS lookups? Some US government agency? Hell NO! We do! Well, at least those of us who run DNS servers do. We do this by means of our vote in the form of the root hints file used in the DNS caching/recursion server. If we point our DNS servers at ICANN sponsered servers, we are effectively designating ICANN to decide for us who runs what TLD, and what TLDs even work.
We could, instead, point our DNS servers at alternative roots. Or we could create our own "." zone and fill in whoever we want for each and every TLDs, add our own, or block ones we don't like.
But really, my point is, if we don't like what ICANN is doing, we should vote with our fingers at the keyboard and switch.
It's been said this can fragment the net. And that's true. But I think the net needs some fragmenting. For example it helps keep RIAA out of my.mp3 (TLD, not file extension) web site. Only those who can figure out how to get to it can get in. I think a little fragmenting is a good thing. I highly doubt very many people will do alternatives to.com,.net, and the country codes (well, at least for most countries). So I won't worry about those being fragmented.
OK, so maybe what you really meant was to see if the domain name on the right side of the @-sign in the sender's email address has an MX record. That's really not useful because most spam uses forged addresses that do, and it's actually valid for a domain name not to have one if it has an A record pointing to the appropriate mail server.
What you want to know is whether the mail really comes from a server it should come from, and SPF is available to do that. If the mail comes from a designated server, you can apply less spam testing, or possibly accept it with no testing at all. However, if the mail comes from somewhere else, you can apply more tests, or maybe even refuse it. But regardless of how you want to do it based on that information, please do not ever send a bounce message back for any reason unless the sending IP address is confirmed as valid for sending mail as from that email address.
Slashdot Mirror
The old unix way of separating things into different directories that are usually on different partitions or drives, has some advantages. Separating state files for the whole machine, state files per user, config files, and binaries (executable and libraries) that are shared between machines on NFS, is a big plus. OTOH, there is definite simplicity and manageability in having everything in one directory for a given application, except for common data more than one application might share.
What would be nice is a way to structure data in a tree such that it appears like like an application oriented structure, where each application has its branch, then underneath that are the different aspects, like binaries here, libraries there, data files, state files, config files, and so on, while still being able to put everything in a specific filesystem where it is manageable in certain ways. One possibility, though messy, is to have an application area under each aspect, like /usr/bin/${appname}/ or /usr/lib/${appname}/. This is often done, especially under /opt/. But Linux does have a way to organize it so you can see it both ways: bind mounts.
What would be needed for security is to make sure that only authorized locations (e.g. the corporate application repository) are accessed. I would never be able to suggest to management to use a system like this unless and until I can thoroughly vet this system for security. And I would never start that process until you can at least tell me you believe it does that. It might be cool from a geek perspective to be able to dynamically download on demand an application from anywhere on the net the author places it, but in business where the geeks have managed to get management to be security aware, such a thing just won't fly. Not only would management want to make sure users won't be running anything someone else might change on the fly, they also want to make sure users are not running anything not approved by management (this in a corporate office setting). If all this can be done restricted to a set of servers on which only approved and vetted applications reside, then you might have something here that is better than the current NFS mounting of / and/or /usr. But if you are designing this for geek coolness factor, it could likely not be suitable for business.
Will it run on Slackware?
So what if I am installing some application which expects its files to be in various places like /usr/share or /etc/foo or wherever? Or is this concept only limited to applications that are aware of ROX?
Competing in this case is fruitless. You can't compete because at the wages that would be paid to do so, you can't afford anything in the US at all. Of course, you could live in luxury in India. Or wait 20 years until the US economy is dragged down to the level of India.
The real problem is not that India has competitors. The problem is that the exchange rate of the US Dollar is such that converting $10000 to Rupees gives someone there plenty enough money to live on very comfortably for a year. It's not a level playing field.
In time, it will level out as more Indians are employed to the point of exhausting its limited supply. India has a few very good schools, comparable to the few top schools in the US, but below that is not much else. In 10 years or so, IIT grads will be demanding higher salaries. I'm already seeing a wee bit of this now from recruiters in India finding it harder to find people to fill job slots.
The problem is, over this time frame, the US gets hurt because there will be far fewer people experienced (because they are not getting experience now) in tech areas, and fare fewer people graduating (because they are turning away from tech majors) with tech degrees. The US companies have always been hurting themselves by insisting on people who are already trained, educated, and/or experienced, and never willing to offer any of that themselves.
Still, the level of outsourcing to India is a small portion of the total unemployment and underemployment in the US. Corporations simply don't have any incentive to do anything other than what making profits gives them, and the US government is failing to provide incentives (like lowering taxes) for specific actions like hiring Americans. The real problem is not that India is able to compete; the real problem is that a US government is too stupid to create the kind of incentives for American businesses to hire more, and hire more Americans.
The benefits can work both ways, either way. With incentives to hire Americans here, the economy prospers because more consumers can buy more products. Business grows, and as the US unemployment diminishes, more employment overseas has to happen, either through US companies going back overseas (scaling back up what they do have), and more imports into the US from overseas. All the arguments about improving the economy work both ways. The issue for the US government is whether to let our own high tech skill base be ravaged in the process.
I build the mission critical programs from source code, and just let the rest be installed as binary packages. I build from source even if I don't need to just to be sure I won't have extra unexpected issues should I ever need to actually make modifications to source and rebuild. I really don't have very many local modifications, but I'm prepared just in case.
Additionally, I do this all on one master machine (with a backup of it kept live on another machine), build binary packages of my own from my source builds, and install those packages on the actual servers. That way I have even more consistency, though at the cost of ultimate optimization. But I think it is better to be able to quickly reinstall a machine, as well as use checksum verifications that there are no trojans.
I use Slackware, but this could be done with most systems, including FreeBSD, Linux (most distributions, including Debian and the RPM based ones), NetBSD, OpenBSD, and even Solaris.
However, using tables for layout is still technically compliant with the standards that specify the table tags. But it's just HTML compliant. The catch with CSS is that the level of "gracefully" degrading isn't satisfactory to lots of people, especially people who come from graphical media like print or video where they have always had explicit control. The only way to overcome this is to do a whole lot better job in two areas. The first is to promote CSS a lot better and more strongly include the requirement to not do the things in HTML that break sites that use CSS. And the other area is to promote better browsers that support CSS correctly, and are not obese. Just because you might have 2 gigabytes of RAM to run Mozilla or whatever does not mean everyone else does. I still have to use Netscape 4 because of these limitations, and some of my friends have even smaller computers than I do. CSS is broken in Netscape 4 and has to be left turned off. I do use Firebird occaisionally, but I can't use it regularly due to the bloat and quite many other bugs (including a form submission security bug). Maybe some day someone with real systems understanding will decide to put together a truly smart modular browser that won't bloat itself beyond what the current web sites it's viewing will need. I can give specific advice, but I won't except directly to the lead developers (because it is a bit complex and would require some back and forth discussion to make sure it is conveyed correctly).
The technology, or its documentation, or its promotion, is at fault when lots of people use it wrong.
If the technology is hard to use then the technology is at fault. For example, a lot of the problems I see are where specific table cells have a specific background color. I don't know that CSS can even specify that, and so maybe it has to be done in HTML, which in turn forces font colors to be done in HTML, too.
Absolutely none of the CSS promoting sites say anything about how to specifically deal with this problem, or with any other specific problem. They describe CSS, tell people it should be used, list the benefits, then leave people hanging with little more than some reference documentation and examples on a few pages that are relevant to real world problems.
Fix that ... e.g. make documentation better ... and maybe things will work out. Now all I need is a decent, reliable, and non-obese browser, that does handle CSS correctly (Netscape 4 doesn't fit the latter requirement, and Mozilla Firebird doesn't fit the former requirement).
The gentoo page spills for me, too. The TLDP page does not. That can't possibly be the same. Just because it might render the same for you doesn't mean it has to for everyone. But if it renders differently even for just a few, then there is at least something different that TLDP did better than gentoo.
Some of the Gentoo page is UNREADABLE. That is because it is attempting (well, succeeding) to change the background color in a table cell without changing the foreground color at the same time. The headings in the tables is where this problem is. Th background is dark blue and the foreground is black. What kind of silliness is that? This is NOT correct graceful degrading; it's screwed up. Of course the fix is easy ... put the necessary font color setting in there. So why are they not doing it? Or do they just not care about older browsers and the people that have damned good reasons to use them.
So in the end, TLDP is more readable to many people.
CSS does NOT always degrade gracefully with HTML ... depending on how well the web developers handle it. If absolutely NO HTML presentation is specified at all, then the site does degrade. Whether that can be called graceful or not is a matter of opinion.
But once a site starts using even the slightest bit of presentation specification in HTML, then things get bad very quickly. For example if a background color is specified in a table cell, it might be in conflict with the default color used for say hyperlinks by the browser. So if the background color is specified, so must every affected font color.
Here is an example of a site, and how it looks in an older browser, where the developers claim they are using "web standards" to make a site, and it degrades horribly. Actually, I would call this site absolute crap. When I communicated with them about it, they simply claimed that it "meets federal accessibility standards". Technically, they are not actually doing web standards correctly. But the point is, there are too many webmasters who are idiots and can never get it right, so I have my doubts about promoting this concept.
Oh, you mean like being paid $1.40 a day (14 hour shift) making US flags?
Here is my idea for a better model for Cable TV. If Congress is going to mandate things, this is what they should mandate.
Make Cable TV simply be a high speed internet where every program provider can make themselves available without having to contract with every individual Cable TV company. The program providers then choose whether to make their programming available free and unencrypted, or encrypted with different payment schemes such as pay per program or pay per month.
The whole idea is to separate the infrastructure provider from the content provider (or as in the case of electricity and gas, the energy provider). With the exception of requirements to carry the basic market over-the air TV stations, Cable TV service should be entirely ala-carte. With digital transmission and encryption like we have now, and computers to interface directly with users and control the access to programming, it should be a very cheap and easy thing to do in the long run.
Is this why EV1 hosts spammers? To make money? Oh yeah, there is money to be made in spam. And all you have to do is just ignore the complaints from other networks that lose money because of the abuse and attacks. Spammers eventually do leave because the IP addresses get majorly blocked, then EV1 can announce that a spammer was terminated an looks good because of that. Of course not everyone falls for it, so many networks have EV1 still blocked. I do, at least until Robert Marsh comes clean about spam.
As for his current situation with SCO ... I'm not considering that to be a "agrees with the anti-SCO crowd" until he puts another million into the fund to defend other companies from SCO that now are under greater risk due to his stupid actions.
Robert Marsh has a lot of expensive cleaning up to do.
And apparently he has the business of hosting spammers, too.
Another best of both worlds scenario. Except that because my network was one of those attacked, I blocked every EV1 address space and those blocks remain until EV1 becomes a clean operating hosting company.
Maybe some of the churn is from being spam-blocked? EV1 has a history of hosting spammers. My network was attacked by one of his spammers for 5 weeks continuously and they would never do a damned thing about it (and Robert Marsh would not take the phone call, either). So they stay blocked until they fix the problem (of not dealing with spam complaints properly), and Robert Marsh has to fix this personally, now.
But his network is NOT rock solid. It has a history of hosting spammers, and at least SMTP is block from that address space on many networks, including mine. The day Robert Marsh announces he will terminate all spammer customers, and accept no more spammers, is the day I look at removing that block. So if he can turn about on the SCO issue, maybe there's hope he can turn about on the spam issue. But his actions will speak more than his words.
Microsoft Office has no threat of viral infection. That's because viral infection is very real. Hell, they ought to remove all doubt and just ship Microsoft Office pre-infected.
Switch to Pike.
I see no reason to even bother using Java anymore. It's been a royal pain where PHP and Python have not been. Open sourcing it may have breathed some life back into it, but ... well ... that didn't happen as I see.
College education really don't help you that much in the knowledge department. It should prepare you for life long on the job learning so you can adapt to all changes (and there will be constant changes). Sadly, while it is the case businesses do want people with genuine on the job experience, they are also unwilling to be the providers of that. Most of them want people that got their on the job experience somewhere else. They seem to also want people that won't learn anything new so they can just shove them out the door in a few years and get someone new and cheaper.
Most positions require experience anyway. Recent grads don't have that. But if you're talking about positions specifically for non-experienced people, then show me those requirements so I can see if they are legitimate or not. I've seen way to many job postings that have unrealistically picky requirements, both of experienced and non-experienced people.
Keep in mind that for every person who does present a level of experience in a skill area, someone did have to hire them once without requiring that skill. They obviously acquired it somewhere (should be obvious where by their resume). Over the past several years, however, businesses have been less willing to hire someone who has to learn some of the needed skills on the job ... and that means the pool of those people who have learned those skills on the job is shrinking.
That's fine for some people. Hell it would be nice if more business executives really did understand the technology better, which would happen if more technical people went the MBA route. But the MBA path isn't right for everyone. Some people just love the technology too much to ever want to do the kind of work an MBA prepares them for. All you really need of these people is that they have a basic understanding of how their technical contribution makes business work. But this does require managers to respect them for their focus on technology (something they could not do as much if they are getting an MBA, too). These are not CTO/CIO jobs I'm talking about here.
That's one of the reasons I left software and went into network and security administration. I happen to love writing software, which is why I don't want to do that at work (it would ruin it for me).
Exactly. Sadly, working for corporations these days tends to rule out doing a job you can love. It's just a job for money. And that is one of the reasons fewer people are going into high tech these days ... because it isn't a career path for money anymore.
Just pray that they don't start outsourcing financial/estate planning to India, as they are already doing to investment analysis and tax preparation. But I'd say your new career path is safe for at least the next 4 or 5 years. Maybe.
You still have to use a valid HTTP "Host:" header with the correct hostname for the site to come up (that's how virtual by name works). And "www.mp3" doesn't work.
There might be a .ogg one, too. But I haven't found any roots for it.
The thing is, these don't need to be in some public root (ICANN, YOUCANN, or FOOCANN); they are intentionally hidden. And if you did stumble across the IP, you'd still have to know what "Host:" header domain name to use to get the site.
Who gives ICANN the right to dictate what TLDs will be usable for our DNS lookups? Some US government agency? Hell NO! We do! Well, at least those of us who run DNS servers do. We do this by means of our vote in the form of the root hints file used in the DNS caching/recursion server. If we point our DNS servers at ICANN sponsered servers, we are effectively designating ICANN to decide for us who runs what TLD, and what TLDs even work.
We could, instead, point our DNS servers at alternative roots. Or we could create our own "." zone and fill in whoever we want for each and every TLDs, add our own, or block ones we don't like.
But really, my point is, if we don't like what ICANN is doing, we should vote with our fingers at the keyboard and switch.
It's been said this can fragment the net. And that's true. But I think the net needs some fragmenting. For example it helps keep RIAA out of my .mp3 (TLD, not file extension) web site. Only those who can figure out how to get to it can get in. I think a little fragmenting is a good thing. I highly doubt very many people will do alternatives to .com, .net, and the country codes (well, at least for most countries). So I won't worry about those being fragmented.
First of all, IP addresses don't have MX records.
OK, so maybe what you really meant was to see if the domain name on the right side of the @-sign in the sender's email address has an MX record. That's really not useful because most spam uses forged addresses that do, and it's actually valid for a domain name not to have one if it has an A record pointing to the appropriate mail server.
What you want to know is whether the mail really comes from a server it should come from, and SPF is available to do that. If the mail comes from a designated server, you can apply less spam testing, or possibly accept it with no testing at all. However, if the mail comes from somewhere else, you can apply more tests, or maybe even refuse it. But regardless of how you want to do it based on that information, please do not ever send a bounce message back for any reason unless the sending IP address is confirmed as valid for sending mail as from that email address.