Disclaimer: this isn't my idea, I got this idea off the January'03 MSDN document browser behavior. Also, although I didn't find the option in mozilla, other tabbed browsers may have this.
When I tab is closed mozilla gives focus to the next tab "physically" in the stack. That is, if you have 5 tabs open, and you open then close a sixth, you'll *always* find yourself staring at the 5th tab.
Mozilla could store a "logical" tab order, or stack. So when I open and close a new tab, the last tab I viewed before that gets focus.
What this means is that if you open a article link from your slashdot tab eg. tab 2 of say 5, and the article opens as the 6th tab; after closing the 6th tab, the 2nd tabs regains focus.
This is simple but very useful. It's almost like the tab focus order acts like the "back" button.
I'm a mozilla guy myself, but I have the January 2003 Microsoft developer network documentation DVD and it's browser uses tabs.
That's interesting because the MSDN document browser application is basically a web browser using the IE engine and shares IE bookmarks. And it hints that microsoft isn't entirely opposed to tabbed browsing.
Links have an "open in new window" right menu item and an "open in separate window" right-click menu, just as mozilla.
I've been wondering if this is a signal of things to come.
I wouldn't recommend instituting a 'women-only' hiring policy, unless you feel like running afoul of Equal Opportunity Employment regulations...
Ah, I actually remembered this from my "human resources" class.
The truth is, it depends on your business. If you run a go-go club, you can hire only women dancers if that's what your clientele (sp.?) wants. It loosely boiled down to whether you are in the entertainment business or not. Interesting case was the one made against "Hooters", you know the restaurant known for it's great um... buffalo wings. Any rate, they where being sued for sex descrimination, since they only hired waitresses. Turns out, all they needed to do was to declare that they were in the entertainment business ( they probably would have had to restructure a bit ) and the suit would have had no merit. Personally though, I agree that they were in violation of the law.
Just think about why we had the problem with antibiotic-resistant staph in the first place: overuse of antibiotics.
Correction misuse of antibiotics, if anything.
Resistant strains may arise simply because the biotics are effective; killing off the susceptible strains and leaving the resistant strains to flourish in environments presumeably with less competition. This has nothing to do with misuse of or overuse.
The biggest threat is when the antibiotic users don't take their treatments to completion.
Keeping this new wonder medicine patented will ensure that it will stay expensive enough that it will only be used when really needed.
Why stop there. There are lots of other medical procedures that would 'benefit' from this arguement. There are tons of viruses ( I'm not going to argue that one ) and bacteria out there. Why not make *all* of their treatments only available to the rich.
However, this really works best only if the aaplication only needs read access! Otherwise, the application's writes need to be referred to the writable master copy, which in this configuration, is unencrypted.
If an app tries to write to the replica, it should recieve a referral. I believe the referral can be an LDAP URI, which means that it can actually specify the encrypted protocol eg. ldaps://master.domain.... .
I don't use referrals though, I write directly to the master from my apps that need write access, in my case. For a large setup, and for redundancy, using referrals are best. The application connects using LDAPS to the master, and regular encrypted LDAP to the local replica.
if you limit your LDAP replica to listening only on the loopback address, then it may not be able to receive updates from the master (unless your LDAP server has a separate replication daemon, i.e., slurpd).
True. In my case, OpenLDAP, does use a separate replication service.
This btw is something I do any chance I can get away with it. If a service does not *have* to listen on the network, I specify localhost as the listening address.
I third option is to compile OpenLDAP with libwrap support and use hosts.allow and friends to specify who connects and who doesn't.
I would recommend using a solution that is going to be compatible with the most other systems without unnecessary complexity. Some recommend using SSH as an easy method.
The thing is, OpenSSH just isn't needed here. That's like running HTTP traffic over OpenSSH for security. Why not run HTTPS instead.
LDAP has LDAPS, ie LDAP over SSL and just about all LDAP servers support this. OpenLDAP does this nicely.
In fact since it's TLS ( next generation SSL ), you don't need a second port for encrypted protocol versus unecrypted. TLS clients can connect to the same port and encryption is negotiated.
Configure your applications to access the local replica using '127.0.0.1' unencrypted. That's the fasted you're going to get without using LDAPI ( LDAP over IPC, eg unix sockets ). At any rate, that's pretty damn fast.
Configure your webserver LDAP replica to only listen on the 'le0' interface, or your localhost interface. For security, also it's usually a good idea to run a firewall.
Configure your replica to replicate with the master server using TLS.
This is for a small setup. Veriations include running a set of LDAP replicas behind your webfarm firewall, etc.
LDAP is a great tool for virtual user setups. We've converted our entire campus network to have all authentication go to LDAP, and neither IT department nor the users have ever been happier.
his way, your replication process can be over the slow SSH tunnel, but your transactions can be fast as they are local.
Your replica idea was spot on I think. Many LDAP administrators don't think that way because they're use to paying huge per server fees. With OpenLDAP that's a moot point so that approach is best.
But you don't need OpenSSH to encrypt your data. You have the much faster native TLS encryption built into the server.
You configure your replicas to replicate using TLS only.
Fortunately, anything anyone is planning to do over LDAP (and particularly OpenLDAP) is extremely unlikely to be performance-critical.
That is simply not true.
I can give you tons of examples of OpenLDAP as *the* mission critical service. It takes care of all the virtual users on all services. IMAP, POP, PAM, HTTP auth, they all access the LDAP system for user authentication.
OpenLDAP is one of the fastest and most reliable LDAP servers ( or services, period ) out there. I have replicas that sit there and do nothing because the masters *never* go down.
In on our campus OpenLDAP was much more reliable than Netscape/iPlanet/SunOne server. We know this because we ran iPlanet for 2 years before making the upgrade.
Not to mention that Bush is under fire for cutting taxes!
Bush is not under fire for just cutting taxes, he's being criticized for who the majority of the cuts are benefiting. Many say the upper class and middle class investors get the large bulk of the savings.
Other complaints include the inappropriate timing of the cuts and whether those cuts can be afforded right now.
altruism - Unselfish concern for the welfare of others; selflessness.
Taking the welfare of others may be a problem when one can not secure their own.
I've heard some people suggest programming is not a real occupation ie. hobby programmers are all we need, and others have reportly suggested programmers should wait tables to make a living while continuing to code in their free time.
I've noticed that funded open source projects do best, while unfunded projects usually don't usually become a force in their market. This goes for open-source and proprietory software equally.
altruism and open source should not be seen as the same thing. Some projects are started because the programmer had an alruistic motives, but unless that/those developer(s) have some source of funding/resources, it becomes very difficult to spend the 3+ hours it takes sometimes to get a decent size project of the ground.
Ask yourself can you, are you, spending that about of time serving the needs of others outside your family in anyway? Without anything in return?
Furthermore, many people (including myself) have cable modems. Upload speeds are "capped" at a fraction of the download speeds.
Maybe it's time to find an ISP that does not put these restrictions on you. Like SpeakEasy for instance.
Unless consumer start rewarding companies that provide services at the consumers terms, and shy away from those ISPs with draconian policies, those caps will only get worst.
Leeching is bad for P2P networks no matter the reason. If everyone acted the same way as the leechers, the value of the P2P would drop considerabley.
Those guys are using persistant server-side applications. Try getting those numbers from a reasonably complex PHP script, even with an opcode cache on such a small box( see my sig. for more info)
Lots of people could use this type performance. I only had a chance to use JSP on one project, a while back. Tomcat was notoriously difficult to install back then. But when it was up, the difference between JSP application server and PHP become apparent. Application servers can make quite the difference.
Just having an application scope for variables saved us a trip the the ldap server per request. PostNUKE, squirellmail, and lots of other large PHP apps could be sped up drastically if some of those features were available in the PHP engine.
The thing that Sendmail did well, address rewriting, is irrelevant for today's net, and can be done more readably in many other ways.
address rewriting is irrelevant... Okay....
Use address rewrite rules to pass messages with addresses that other process expect not ones that the user provides. Eg. user setups 'user@mail.domain.tld' and cyrus only knows 'user@domain.tld'.
Address rewrites are useful for rewriting addresses when clients pull a random domain from system configuration, and sends that out in a message.
rewrites are handy when to split users between servers, etc. There are lots more cases.
Even a Perl script is much more readable:-) Installing one as a Postfix transport is trivial.
Yeah, you do that. That's just 'screaming' performance, Bruce:)
Sendmail is so bloated that it apparently takes books to get decent performance out of it.
Seriously, a lot of people just don't understand sendmail and how difficult mail routing is. Sendmail is bloated because it tries to solve a very complex problem in an extremely diverse environment.
Postfix was so easy to use, that I went from reading the man pages to a working system in one hour.
Postfix and sendmail are going in two different directions. Sendmail == configurability, postfix == ease of use.
Sendmail is fast. Not because there's a book on performance tuning does it mean the software is slow. Sendmail is probably as fast as anything else out there.
Sendmail is configurable to no end. I run a sendmail setup with virtual domains, lots of address rewrite rules, spam filters. All my virtual users, and alias maps are in LDAP. I modify all my sendmail virtual users, virtual domains, aliases, etc through a PHP website. Try that with other mailers. They support LDAP, but they are no where as configurable as sendmail.
Then try running these mailers on Linux, bsd, solaris, windows, AIX, etc., etc. No OS lock-in.
Sendmail is well documented.
The prevaling mood on/. these days seems to be if it's not brand new, get rid of it. That's unfortunate.
One of the nice things about knowing and using a number of languages is that you get to pick the right tool for the right job. People like you, Kunta Kinte, seem to believe it's a good thing to limit tools;
You missed my point entirely.
I am not saying one shouldn't learn as many languages as they can. We should. Neither am I saying you should use one tool for all situations. Obviously not.
My point was to be careful how you classify your knowledge.
Lots of people to to come off as an Java expert because they 'did a class'. When I did compiler theory class and we each wrote a compiler, from scratch, with the lexer and parser written in Flex and Bison. Does that make me an expert in Flex/Bison? Don't think so. That qualifies as familiar in my view, even together with a few other encounters in miscellaneous projects.
How many environments do you work in? I'm equally at home in the embedded space, personal-computer applications, Web applications, secure e-commerce applications, network stuff, and man-rated programming. Each area has its own set of tools -- why shouldn't I mention them as I'm versed in using them?
My guess is, unless you are in the top 10 percentile of programmers out there, I'd say you can easily shave a few of those off. Geeks often have an overly rosey view of themselves.
But that's just my guess. I don't know you.
PS. Try not to take my view as a personal attack. It's just an opinion.
This isn't offtopic, it's research into the right buzzwords to get a job. Do you know even one serious programmer who can't code in any language if presented a pocket reference guide? Citing specific languages in a job search is all about buzzwords for resumes.
I strongly disagree with that approach.
I've picked up a working knowledge of many languages over the years but I'm not to say that I am an expert, or even proficient.
Expertise in a language implies you know the compiler and runtime environment very, very well. It also implies that you know the common pitfalls, strengths concerning the langauage and you know how to deal with them.
It's all vague "proficient", "expert", "knows". I'd say what you discribed wouldn't pass for more than "familar", ie. "familar with language x"
I am very wary of people that list 20 different languages on their resume, or suggest that they know these languages otherwise. Not that I'm in a position to make hiring decisions right now though.
I noticed that X Windows critics never give any solid proof to their claims. there is no 'my app runs x fast on framebuffer but yx fast under X'
Those posts are getting tired. There should be a faq somewhere.
X is not a memory hog. The protocol is lean, think of when it was designed. It couldn't afford to be a memory hog. X can be 'fast'. X is very modular. X runs on embedded systems that have very little resources, and I mean *very* little.
Comparing the X network transparency to RDP is like comparing apples to oranges. Frambuffer based transparency eg. RDP work well on low bandwidth situations but push all the load on the server since the entire application and all the rendering is done on the server. This is a terrible design in thin client networks, and why citrix et. all take so much resources to deploy. I've seen Solaris boxes push a ridiculus amount of concurrent sessions while MS terminal services halt at a fraction of that load. It's not that MS did a bad job, it's just that the two approaches have their strong suits.
The bottom line is learn X before you diss it ( someone else said that ).
It's true, most of the H1B's I know have **mad skillz** with the ladies. I don't know if they take classes for that shit over there but man, those guys get laid 5,6, hell 100 times more often than I do.
And by better looking chicks!
It seems they're willing to put more effort into getting nookie. And that ain't right.
There I am putting down some smooth lyrics on a honey, chilling on campus, you know the deal. When out of nowhere this arabian guy drives by with a sup'ed up navigator with oversized chromes.
oversized fuckin' chromes! How the hell am I suppose to compete with oversized fuckin' chromes!!!
I'd say to hell with the job market, sharing with a roommate aint bad. As long as he pays the rent and stays out of my room at night ( I still don't buy that sleep walkin' story ). Just keep those guys away from our freakin' women.
PS.: TechsUnite and the rest of you guys sound as ridiculus as I do.
Slashdot Mirror
Disclaimer: this isn't my idea, I got this idea off the January'03 MSDN document browser behavior. Also, although I didn't find the option in mozilla, other tabbed browsers may have this.
When I tab is closed mozilla gives focus to the next tab "physically" in the stack. That is, if you have 5 tabs open, and you open then close a sixth, you'll *always* find yourself staring at the 5th tab.
Mozilla could store a "logical" tab order, or stack. So when I open and close a new tab, the last tab I viewed before that gets focus.
What this means is that if you open a article link from your slashdot tab eg. tab 2 of say 5, and the article opens as the 6th tab; after closing the 6th tab, the 2nd tabs regains focus.
This is simple but very useful. It's almost like the tab focus order acts like the "back" button.
That's interesting because the MSDN document browser application is basically a web browser using the IE engine and shares IE bookmarks. And it hints that microsoft isn't entirely opposed to tabbed browsing.
Links have an "open in new window" right menu item and an "open in separate window" right-click menu, just as mozilla.
I've been wondering if this is a signal of things to come.
If you plan to start a IT consulting firm where the techs give lap-dances while configuring your routers, I think you're in the clear.
Ah, I actually remembered this from my "human resources" class.
The truth is, it depends on your business. If you run a go-go club, you can hire only women dancers if that's what your clientele (sp.?) wants. It loosely boiled down to whether you are in the entertainment business or not. Interesting case was the one made against "Hooters", you know the restaurant known for it's great um... buffalo wings. Any rate, they where being sued for sex descrimination, since they only hired waitresses. Turns out, all they needed to do was to declare that they were in the entertainment business ( they probably would have had to restructure a bit ) and the suit would have had no merit. Personally though, I agree that they were in violation of the law.
Correction misuse of antibiotics, if anything.
Resistant strains may arise simply because the biotics are effective; killing off the susceptible strains and leaving the resistant strains to flourish in environments presumeably with less competition. This has nothing to do with misuse of or overuse.
The biggest threat is when the antibiotic users don't take their treatments to completion.
Keeping this new wonder medicine patented will ensure that it will stay expensive enough that it will only be used when really needed.
Why stop there. There are lots of other medical procedures that would 'benefit' from this arguement. There are tons of viruses ( I'm not going to argue that one ) and bacteria out there. Why not make *all* of their treatments only available to the rich.
I thought that was Henry Ford. You can have a car in any color you want, as long as it's black.
Why, thank you. :)
However, this really works best only if the aaplication only needs read access! Otherwise, the application's writes need to be referred to the writable master copy, which in this configuration, is unencrypted.
If an app tries to write to the replica, it should recieve a referral. I believe the referral can be an LDAP URI, which means that it can actually specify the encrypted protocol eg. ldaps://master.domain.... .
I don't use referrals though, I write directly to the master from my apps that need write access, in my case. For a large setup, and for redundancy, using referrals are best. The application connects using LDAPS to the master, and regular encrypted LDAP to the local replica.
if you limit your LDAP replica to listening only on the loopback address, then it may not be able to receive updates from the master (unless your LDAP server has a separate replication daemon, i.e., slurpd).
True. In my case, OpenLDAP, does use a separate replication service.
This btw is something I do any chance I can get away with it. If a service does not *have* to listen on the network, I specify localhost as the listening address.
I third option is to compile OpenLDAP with libwrap support and use hosts.allow and friends to specify who connects and who doesn't.
Ok, sorry about that.
But I still disagree with this statement. LDAP servers are often optimized for read performance and for storing small "records" or entries.
Because of this they perform better than SQL databases at what they do.
Honestly, I don't know of a faster network datastore one could find.
The thing is, OpenSSH just isn't needed here. That's like running HTTP traffic over OpenSSH for security. Why not run HTTPS instead.
LDAP has LDAPS, ie LDAP over SSL and just about all LDAP servers support this. OpenLDAP does this nicely.
In fact since it's TLS ( next generation SSL ), you don't need a second port for encrypted protocol versus unecrypted. TLS clients can connect to the same port and encryption is negotiated.
This is for a small setup. Veriations include running a set of LDAP replicas behind your webfarm firewall, etc.
LDAP is a great tool for virtual user setups. We've converted our entire campus network to have all authentication go to LDAP, and neither IT department nor the users have ever been happier.
Good luck
Your replica idea was spot on I think. Many LDAP administrators don't think that way because they're use to paying huge per server fees. With OpenLDAP that's a moot point so that approach is best.
But you don't need OpenSSH to encrypt your data. You have the much faster native TLS encryption built into the server.
You configure your replicas to replicate using TLS only.
That is simply not true.
I can give you tons of examples of OpenLDAP as *the* mission critical service. It takes care of all the virtual users on all services. IMAP, POP, PAM, HTTP auth, they all access the LDAP system for user authentication.
OpenLDAP is one of the fastest and most reliable LDAP servers ( or services, period ) out there. I have replicas that sit there and do nothing because the masters *never* go down.
In on our campus OpenLDAP was much more reliable than Netscape/iPlanet/SunOne server. We know this because we ran iPlanet for 2 years before making the upgrade.
Okay, mostly for porn.
But, instead of persuing global conventions that would deter this from happening, effort is under way to legitimize it.
This is so sad. I suspect many future generations may suffer for some of the mistakes being made today.
Bush is not under fire for just cutting taxes, he's being criticized for who the majority of the cuts are benefiting. Many say the upper class and middle class investors get the large bulk of the savings.
Other complaints include the inappropriate timing of the cuts and whether those cuts can be afforded right now.
Officials: Government close to hitting debt ceiling
Let the children pay
Taking the welfare of others may be a problem when one can not secure their own.
I've heard some people suggest programming is not a real occupation ie. hobby programmers are all we need, and others have reportly suggested programmers should wait tables to make a living while continuing to code in their free time.
I've noticed that funded open source projects do best, while unfunded projects usually don't usually become a force in their market. This goes for open-source and proprietory software equally.
altruism and open source should not be seen as the same thing. Some projects are started because the programmer had an alruistic motives, but unless that/those developer(s) have some source of funding/resources, it becomes very difficult to spend the 3+ hours it takes sometimes to get a decent size project of the ground.
Ask yourself can you, are you, spending that about of time serving the needs of others outside your family in anyway? Without anything in return?
Lotus notes
oracle collaboration
Bynari InsightServer.
There are others.
Maybe it's time to find an ISP that does not put these restrictions on you. Like SpeakEasy for instance.
Unless consumer start rewarding companies that provide services at the consumers terms, and shy away from those ISPs with draconian policies, those caps will only get worst.
Leeching is bad for P2P networks no matter the reason. If everyone acted the same way as the leechers, the value of the P2P would drop considerabley.
Lots of people could use this type performance. I only had a chance to use JSP on one project, a while back. Tomcat was notoriously difficult to install back then. But when it was up, the difference between JSP application server and PHP become apparent. Application servers can make quite the difference.
Just having an application scope for variables saved us a trip the the ldap server per request. PostNUKE, squirellmail, and lots of other large PHP apps could be sped up drastically if some of those features were available in the PHP engine.
Next thing you know, they'll be making money.
I think these are the innovations that the linux distros need even more than new drivers, other technical advances.
The thing that Sendmail did well, address rewriting, is irrelevant for today's net, and can be done more readably in many other ways.
address rewriting is irrelevant... Okay....
Use address rewrite rules to pass messages with addresses that other process expect not ones that the user provides. Eg. user setups 'user@mail.domain.tld' and cyrus only knows 'user@domain.tld'.
Address rewrites are useful for rewriting addresses when clients pull a random domain from system configuration, and sends that out in a message.
rewrites are handy when to split users between servers, etc. There are lots more cases.
Even a Perl script is much more readable :-) Installing one as a Postfix transport is trivial.
Yeah, you do that. That's just 'screaming' performance, Bruce :)
...to repeat it.
Um, wait. That didn't come out right.
Sendmail is so bloated that it apparently takes books to get decent performance out of it.
Seriously, a lot of people just don't understand sendmail and how difficult mail routing is. Sendmail is bloated because it tries to solve a very complex problem in an extremely diverse environment.
Postfix was so easy to use, that I went from reading the man pages to a working system in one hour.
Postfix and sendmail are going in two different directions. Sendmail == configurability, postfix == ease of use.
Sendmail is fast. Not because there's a book on performance tuning does it mean the software is slow. Sendmail is probably as fast as anything else out there.
Sendmail is configurable to no end. I run a sendmail setup with virtual domains, lots of address rewrite rules, spam filters. All my virtual users, and alias maps are in LDAP. I modify all my sendmail virtual users, virtual domains, aliases, etc through a PHP website. Try that with other mailers. They support LDAP, but they are no where as configurable as sendmail.
Then try running these mailers on Linux, bsd, solaris, windows, AIX, etc., etc. No OS lock-in.
Sendmail is well documented.
The prevaling mood on /. these days seems to be if it's not brand new, get rid of it. That's unfortunate.
You missed my point entirely.
I am not saying one shouldn't learn as many languages as they can. We should. Neither am I saying you should use one tool for all situations. Obviously not.
My point was to be careful how you classify your knowledge.
Lots of people to to come off as an Java expert because they 'did a class'. When I did compiler theory class and we each wrote a compiler, from scratch, with the lexer and parser written in Flex and Bison. Does that make me an expert in Flex/Bison? Don't think so. That qualifies as familiar in my view, even together with a few other encounters in miscellaneous projects.
How many environments do you work in? I'm equally at home in the embedded space, personal-computer applications, Web applications, secure e-commerce applications, network stuff, and man-rated programming. Each area has its own set of tools -- why shouldn't I mention them as I'm versed in using them?
My guess is, unless you are in the top 10 percentile of programmers out there, I'd say you can easily shave a few of those off. Geeks often have an overly rosey view of themselves.
But that's just my guess. I don't know you.
PS. Try not to take my view as a personal attack. It's just an opinion.
I strongly disagree with that approach.
I've picked up a working knowledge of many languages over the years but I'm not to say that I am an expert, or even proficient.
Expertise in a language implies you know the compiler and runtime environment very, very well. It also implies that you know the common pitfalls, strengths concerning the langauage and you know how to deal with them.
It's all vague "proficient", "expert", "knows". I'd say what you discribed wouldn't pass for more than "familar", ie. "familar with language x"
I am very wary of people that list 20 different languages on their resume, or suggest that they know these languages otherwise. Not that I'm in a position to make hiring decisions right now though.
Those posts are getting tired. There should be a faq somewhere.
X is not a memory hog. The protocol is lean, think of when it was designed. It couldn't afford to be a memory hog. X can be 'fast'. X is very modular. X runs on embedded systems that have very little resources, and I mean *very* little.
Comparing the X network transparency to RDP is like comparing apples to oranges. Frambuffer based transparency eg. RDP work well on low bandwidth situations but push all the load on the server since the entire application and all the rendering is done on the server. This is a terrible design in thin client networks, and why citrix et. all take so much resources to deploy. I've seen Solaris boxes push a ridiculus amount of concurrent sessions while MS terminal services halt at a fraction of that load. It's not that MS did a bad job, it's just that the two approaches have their strong suits.
The bottom line is learn X before you diss it ( someone else said that ).
...cause X rocks!!!
It's true, most of the H1B's I know have **mad skillz** with the ladies. I don't know if they take classes for that shit over there but man, those guys get laid 5,6, hell 100 times more often than I do.
And by better looking chicks!
It seems they're willing to put more effort into getting nookie. And that ain't right.
There I am putting down some smooth lyrics on a honey, chilling on campus, you know the deal. When out of nowhere this arabian guy drives by with a sup'ed up navigator with oversized chromes.
oversized fuckin' chromes! How the hell am I suppose to compete with oversized fuckin' chromes!!!
I'd say to hell with the job market, sharing with a roommate aint bad. As long as he pays the rent and stays out of my room at night ( I still don't buy that sleep walkin' story ). Just keep those guys away from our freakin' women.
PS.: TechsUnite and the rest of you guys sound as ridiculus as I do.