What? You cannot seriously compare PHP/Perl (even mod-perl) with Java/J2EE (as an application server) for large projects. PHP and Perl work great for small web apps, blogs, etc. But for maintainability, Java/J2EE are great for large projects.
As far as I know, Python doesn't have an application server container (ala Tomcat style), so that's out. Ruby (from what I've read; never used it) seems promising. Java has been around for years, so software companies that have been using Java for years have hopefully learned from mistakes. They'd have to re-learn from mistakes if starting with Ruby.
Of course, bad design and bad implementations can make a large project done in Java/J2EE fail miserably. But that goes for any language.
A main reason I'm using Gaim is that I never see any emoticons - especially custom ones. I wish the protocols had an option to display plain-text emoticons when I send them.
Side note: Outlook should send ":)" instead of the freaking "J". Man that bugs me.
They can load-balance at so many levels (like DNS, TCP/IP, Application level, etc.) that this would be a ridiculous scheme to do so.
It seems like whoever implemented this was trying to do some silly IP logging scheme. (Which is just as ridiculous.) S/he must have not known how to grab IPs from the initial web request.
There was a good Futurama episode (season four I think) where the ship went down to the bottom of an ocean. And since it was built for space, it could not withstand the pressure at depth. Seems they had no pressure problems after returning to the surface though.
yes! Anyone know of any decent removable flash-based compressed audio players? Compact flash would be sweet -- 2-3 years later we'd have 5GB+ removable media; for now we could settle for 2GB.
I'd say.com +.net +.org (the gTLDs) are about 80% currently. However, existing ccTLDs are growing in use around the world, in addition to new ccTLDs like.EU. The gTLD namespace has somewhat stagnated.. yeah, there are a lot of new domain purchases, but there almost as many expirations/deletions.
For future growth, the hierarchical model makes sense. As asia and europe grow in the internet market, ccTLDs (like.cn, or the IDN equivalents) will grow.
Granted, semantically, the gTLDs have no use, but the ccTLDs do. Most of the motivation for the specialized TLDs (like.pro,.xxx,.tel,.mobi, etc.) are to make money, as the registy operators and the registries know that the large corporate clients will purchase domains.
Because our 13 root servers would then need to store every domain name in existance (NS and A records). It entirely defeats the purpose of a hierarchical database, which the Domain Name System is.
Well, assuming this OpenID thing is really great and wonderful and doesn't make the baby jesus cry, then perhaps a lot of small sites will use it. And if a lot of small sites are using it, it might trickle up to a decent amount of medium sites, which might get noticed by a few large sites.
No one liked Passport so that's why it didn't get used. This is a different idea which has a slim, but possible, chance of success.. even on large sites.
The other considered .tel proposal
on
.tel Coming Soon
·
· Score: 1
the other proposed usages of.tel had a potential to cause damage in addition to being useless
I can see how there could be some confusion with the numbering systems in the world if we had the number-based.tel TLD. But, it could at least be potentially useful. I see no value in a name-based.tel except that the registry operator, accredited registrars, and ICANN will all make money.
Will the name-based.tel be for the good of the internet? Or in the internet's common interest? Probably not. Might the number-based.tel aid in VOIP or 'bridge the gap between phone networks and computer networks'? Maybe.
What if all browsers / proxies displayed themselves as IE? Then these stats would be useless as well as the unwanted javascript that checks for the user agent.
But water is not wet -- only something saturated by water is wet. Spill water on your shirt, your shirt is wet. Pour water on some soil, the soil is wet. Pour water into more water, it's still water.
I guess this depends if you want one site to be a failover or if you want them to be semi-load-balanced.
First, you don't have to worry about TTLs for the MX records, multiple records with priorities (same number or otherwise) will take care of themselves.
For web, if you want load balancing, you can simply create two A records for www.domain -- one IP for server X, one IP for server Y. I'm not entirely sure if both records are returned to the client (browser) or if the ISP will only cache one. Assuming they cache both records and the client gets both A records, it should work in case either server were to fail. (Client should attempt 2nd IP address if first fails).
For simple failover (say server X is beefy, server Y is only for emergencies) then a low TTL should work -- assuming: 1) you can handle the DNS requests and 2) that the Net will honor your TTL.
Also, you only need the low TTL for your www A records.. your SOA, NS, MX, TXT, etc. should all have a reasonable TTL (>3 hours). There can also be many layers of DNS cache -- the first may be the ISPs resolver, the second could be the ISPs forwarder, and the third level might be the client. So the perceived TTL in this case is 3 * the A record TTL.
Remember everyone, Time To Live is based on the honor system. If your neighbor doesn't wish to honor your TTL, too bad.
Exactly. It seems like a way to have a "text" file that is easily parsed (all the XML info -- in this case possibly a description, comments, image meta-data, etc.), yet binary info (a jpeg compressed image) fits along-side for when you want it. One file with all the goodies.
How this is different than simply base64 encoding the image inside a tag is yet to be seen. Perhaps because it's a standard?
You don't have to do it every time, just once per folder.. After that, the 'check mail' button will look for new mail in all of those folders. I have server-side rules that put stuff into folders that I do not want to be notified about. So I just 'check for new messages' only on certain folders. If all your rules are client-side, then this does not apply.
Not sure about server software (try freshmeat), but the IMAP protocol is excellent, IMO. Whereas POP is terrible.
I would love to have all sorts of information stored on IMAP-like servers.. like bookmarks, accessible anywhere, since they stay on the server.
Actually, you can right-click on each folder >> properties >> check this folder for new messages.
Pretty simple. I use it to connect to echange via IMAPS, I write my rules with outlook so they sit on the server, then they get filtered into the appropriate folders.
-- The part about having to download the whole message with the attachment is painful. But I've not used any other imap client that doesn't work this way. Pine would take forever to load an attached message over imap too. Perhaps it is the imap implementation by the exchange server?
It takes me a lot shorter to install Win98 on a box and that includes saving any or all documents.
1.5 hours tops.
I agree. Besides, you need to reinstall Win98 every 6 months anyway. I had a dual-boot machine a few years ago. I hadn't booted to Win98 in about 5 months. (The semester ended so it was time to play some old games.) I hadn't changed any hardware, but on boot, Windows "forgot" about my video drivers and network card. And that was 5 months not even running windows!
What? It seems to me that anyone who cannot install SP1 cannot install SP2. Are all the cracked machines able to install SP1?
From parent's link:
The same users that were blocked from installing SP1 - those that have used a small set of legacy pirated product keys - will be blocked from installing SP2.
I believe Gary Schare is saying the "broadest number of people [who] can install SP2" are valid users and "pirates [who] have moved on to other keys which we are not blocking." So.. Are some non-purchased keys (that work with SP1 and updates) not blocked? If not, this means the [many?] non-purchased copies of XP will still be less secure.
The disk is still the slowest part of your system (well, after the network and external devices). Fast RAM is nice, but only after your data has been loaded from disk into memory! So the biggest boost in performance visible to the user would be a faster disk -- try a 10k rpm SATA drive for your OS/games/apps, and larger/cheaper 7200 rpm drives for data storage.
Anyone else getting a little tired of the constant hype of new consoles? This is getting a bit rediculous. A few years ago it was all hype about xbox 1, PS2, and gamecube. A month after those consoles hit the market, the media turned to hyping the next gen consoles.
Wake me up when the next console is ready to be sold, give me full tech specs with game screenshots, and I might give it a nod.
If the world moves too fast on these, small to mid-sized developers won't have the time to develop a game for the current console -- they'll always be anticipating the next generation and games will cease to be polished.
Slashdot Mirror
What? You cannot seriously compare PHP/Perl (even mod-perl) with Java/J2EE (as an application server) for large projects. PHP and Perl work great for small web apps, blogs, etc. But for maintainability, Java/J2EE are great for large projects.
As far as I know, Python doesn't have an application server container (ala Tomcat style), so that's out. Ruby (from what I've read; never used it) seems promising. Java has been around for years, so software companies that have been using Java for years have hopefully learned from mistakes. They'd have to re-learn from mistakes if starting with Ruby.
Of course, bad design and bad implementations can make a large project done in Java/J2EE fail miserably. But that goes for any language.
A main reason I'm using Gaim is that I never see any emoticons - especially custom ones. I wish the protocols had an option to display plain-text emoticons when I send them.
Side note: Outlook should send ":)" instead of the freaking "J". Man that bugs me.
They can load-balance at so many levels (like DNS, TCP/IP, Application level, etc.) that this would be a ridiculous scheme to do so.
It seems like whoever implemented this was trying to do some silly IP logging scheme. (Which is just as ridiculous.) S/he must have not known how to grab IPs from the initial web request.
Now, I do like the myriad of K* applications. If only they were more pleasing to the eyes.
Just like they did with SVCD, the Chinese version of a high(er) definition digital versatile disc could be 'teh winnar!' here.
Sometimes in chess, you'll let a pawn be sacrificed to set up a checkmate.
Sometimes? If all it takes is a pawn sacrifice, I would say *always*.
But then again, I haven't been following this game move by move.. I'm just waiting for the Deep Blue victory.There was a good Futurama episode (season four I think) where the ship went down to the bottom of an ocean. And since it was built for space, it could not withstand the pressure at depth. Seems they had no pressure problems after returning to the surface though.
s/post/'news' story/g
...64kb [is] listenable on WMA though.
hah! - maybe listenable only by those people who are near deaf from listening to headphones too lound. The 160kb+ WMA is listenable. Barely.yes! Anyone know of any decent removable flash-based compressed audio players? Compact flash would be sweet -- 2-3 years later we'd have 5GB+ removable media; for now we could settle for 2GB.
For future growth, the hierarchical model makes sense. As asia and europe grow in the internet market, ccTLDs (like .cn, or the IDN equivalents) will grow.
Granted, semantically, the gTLDs have no use, but the ccTLDs do. Most of the motivation for the specialized TLDs (like .pro, .xxx, .tel, .mobi, etc.) are to make money, as the registy operators and the registries know that the large corporate clients will purchase domains.
Because our 13 root servers would then need to store every domain name in existance (NS and A records). It entirely defeats the purpose of a hierarchical database, which the Domain Name System is.
No one liked Passport so that's why it didn't get used. This is a different idea which has a slim, but possible, chance of success.. even on large sites.
First, these are the two .TEL applications:
I can see how there could be some confusion with the numbering systems in the world if we had the number-based .tel TLD. But, it could at least be potentially useful. I see no value in a name-based .tel except that the registry operator, accredited registrars, and ICANN will all make money.
Will the name-based .tel be for the good of the internet? Or in the internet's common interest? Probably not. Might the number-based .tel aid in VOIP or 'bridge the gap between phone networks and computer networks'? Maybe.
Related: Firefox extension - user agent switcher
But water is not wet -- only something saturated by water is wet. Spill water on your shirt, your shirt is wet. Pour water on some soil, the soil is wet. Pour water into more water, it's still water.
First, you don't have to worry about TTLs for the MX records, multiple records with priorities (same number or otherwise) will take care of themselves.
For web, if you want load balancing, you can simply create two A records for www.domain -- one IP for server X, one IP for server Y. I'm not entirely sure if both records are returned to the client (browser) or if the ISP will only cache one. Assuming they cache both records and the client gets both A records, it should work in case either server were to fail. (Client should attempt 2nd IP address if first fails).
For simple failover (say server X is beefy, server Y is only for emergencies) then a low TTL should work -- assuming:
1) you can handle the DNS requests and
2) that the Net will honor your TTL.
Also, you only need the low TTL for your www A records.. your SOA, NS, MX, TXT, etc. should all have a reasonable TTL (>3 hours). There can also be many layers of DNS cache -- the first may be the ISPs resolver, the second could be the ISPs forwarder, and the third level might be the client. So the perceived TTL in this case is 3 * the A record TTL.
Remember everyone, Time To Live is based on the honor system. If your neighbor doesn't wish to honor your TTL, too bad.
Exactly. It seems like a way to have a "text" file that is easily parsed (all the XML info -- in this case possibly a description, comments, image meta-data, etc.), yet binary info (a jpeg compressed image) fits along-side for when you want it. One file with all the goodies.
How this is different than simply base64 encoding the image inside a tag is yet to be seen. Perhaps because it's a standard?
Not sure about server software (try freshmeat), but the IMAP protocol is excellent, IMO. Whereas POP is terrible.
I would love to have all sorts of information stored on IMAP-like servers.. like bookmarks, accessible anywhere, since they stay on the server.
Actually, you can right-click on each folder >> properties >> check this folder for new messages.
Pretty simple. I use it to connect to echange via IMAPS, I write my rules with outlook so they sit on the server, then they get filtered into the appropriate folders.
-- The part about having to download the whole message with the attachment is painful. But I've not used any other imap client that doesn't work this way. Pine would take forever to load an attached message over imap too. Perhaps it is the imap implementation by the exchange server?
Dude, call Guinness for a record!
I agree. Besides, you need to reinstall Win98 every 6 months anyway. I had a dual-boot machine a few years ago. I hadn't booted to Win98 in about 5 months. (The semester ended so it was time to play some old games.) I hadn't changed any hardware, but on boot, Windows "forgot" about my video drivers and network card. And that was 5 months not even running windows!
From parent's link:
I believe Gary Schare is saying the "broadest number of people [who] can install SP2" are valid users and "pirates [who] have moved on to other keys which we are not blocking." So.. Are some non-purchased keys (that work with SP1 and updates) not blocked? If not, this means the [many?] non-purchased copies of XP will still be less secure.
The disk is still the slowest part of your system (well, after the network and external devices). Fast RAM is nice, but only after your data has been loaded from disk into memory! So the biggest boost in performance visible to the user would be a faster disk -- try a 10k rpm SATA drive for your OS/games/apps, and larger/cheaper 7200 rpm drives for data storage.
Wake me up when the next console is ready to be sold, give me full tech specs with game screenshots, and I might give it a nod.
If the world moves too fast on these, small to mid-sized developers won't have the time to develop a game for the current console -- they'll always be anticipating the next generation and games will cease to be polished.