So I wouldn't be too smug about the high-end users, because they are paying *more* for Android than they would pay for iPhone.
I pay under $20/mo for my Palm Pre. It would be under $30/mo for an Evo (have to include the 4G tax). That's with actual unlimited 4G data (limited only by your throughput, not 5GB), unlimited text, unlimited GPS, mobile-to-mobile calling (on any network), and whatever else they include that I don't care about.
I don't know how much an iphone costs, but I bet it's a lot fucking more than that. For me, switching to iPhone would be prohibitively expensive. It's a great device, but your argument that it's cheap is bogus.
The reason I didn't consider an upgrade to the Evo is that the device is simply too big.
I'm not really invested in one platform over another. If Sprint released a CDMA iPhone, I'd consider it. Not sure how I feel about it lacking a keyboard, but I'd give it a shot. And if it doesn't sync with google like WebOS and Android, then it's a nonstarter for me. All my shit's in google right now.
I try not to ask "what is this?" or "define this" type of questions, but I try to ask problem solving questions to see their problem solving skills.
Yeah, that's why I like to do both. But I don't typically ask "define this", per se. I'll ask "What's the difference between?" instead. That way, if they don't have a solid understanding of something on their resume, I'll find out immediately.
Example candidate whom I did not hire had DOM and SAX on his resume: Me: What is the difference between DOM and SAX? Candidate: They are both for Java XML programming. Me: Yes, but they each approach that task in a different way. Can you please elaborate on that? Candidate:......... Me: Would you like me to do you the favor of crossing those buzzwords off your resume so we can move on? Candidate:... uh, yeah.
Needless to say, I did not select that candidate.
Free advice to all your job seekers out there: Don't put anything on your resume unless you are prepared to discuss that single topic for the entire length of the interview. It probably won't happen, but if that one buzzword is what I'm hiring you for, you're going to look like an idiot if the best you can do is expand the acronym!
The whole idea of LAMP is that it's an easy-to-learn, easy-to-deploy stack. Any competent developer should be able to learn this, quickly. Even if you could assign a "LAMP Competence Metric" (say a 0-10 scale) to a person, a competent developer who is a 2 in LAMP will be a 9 much more quickly than an incompetent developer who is currently a 6.
When I hire coders, I like to see how quickly they can understand a system via standard UML architecture diagrams. I like to see if they can implement a basic logic task using their choice of language, and I want that implementation to be straightforward, and I want it to use the coder's chosen language's foundation libraries instead of reinventing the wheel.
I also like to get a feel for how experienced they are in working with any of the various system development processes. (Hint: "What's that?" is a bad answer. Cowboy Coders are unwelcome at my company.)
Regarding testing basic computer science concepts, I like to ask candidates to differentiate one thing from another. For example, given your list of concepts ("BNF, data normalization, OOP, MVC"), I'd ask:
I'd show them a BNF for something simple and ask them to interpret it. (Honestly though, I haven't used the term BNF since college. Not sure why you care.)
What is the difference between a normalized and denormalized schema? In what situations might you use each?
What is the difference between OOP and procedural programming?
Can you please describe MVC and what problems it solves?
But apart from that you really, really want to look at what causes connection problems - Apple is likely far less guilty than you think.
I'm not an RF engineer by any stretch of the imagination, but I'll offer you this anecdote:
My house is in a little bit of a tricky zone, from a cellular reception perspective. Verizon is the only carrier that gets perfect reception; Sprint gets reasonable reception, but can roam on Verizon; AT&T gets weak, but usable signal; T-Mo gets nothing.
Of my friends on AT&T, the ones with blackberries can use their phones just fine at my house (T-Mo blackberries do not work, for some reason). On the other hand, iPhones (3G and 3Gs) just do not work. At all. Not inside. Not outside. Not on the roof. Not in a tree. Not with green eggs and ham.
If you were to stop by my house, your iPhone would be reduced to a bulky iPod Touch.
I'm sure there's some logical explanation from an RF perspective, but at least now you know why I tend to blame Apple first.:P
This isn't really about cheating, and in most states, cheating isn't relevant in a divorce settlement. Welcome to the wonderful world of no-fault divorce.
Where Facebook can really cause problems is in custody battles, or in lying to a judge. Consider how this might work.
gavron: No, your honor, I don't have a drinking problem. I'll enjoy a glass of wine with a meal from time to time, but I don't have a drinking problem. Meanwhile, on FB: gavron is so fucking hungover right now and doesn't remember a goddamn thing from last night. It's just as well, as he really doesn't want to know if the new urine stain on his carpet was created by him or his miniature schnauzer.
gavron: I can't afford such a high child support payment. The economy is so bad, I was forced to accept a job delivering pizzas. I can't find work in my field, despite trying as best I can. Meanwhile, on FB: gavron just got a great new job at Initech as a middle manager. He is so excited that he doesn't have to deliver any more goddamn pizzas!
Even if you are not so stupid as to hang yourself like that on facebook, you can't be sure that your friends will be equally clever. ([gavon's drinking buddy] wants to congratulate gavon on his new job. Don't forget those TPS reports, man! Happy hour tonight?)
The problem is that divorce cases can get complicated, and it's very easy to let something slip. It's one thing to say the wrong thing; it's quite another to post the wrong thing publicly.
In general, infidelity is a non-issue in divorce cases. I've been divorced twice, both times because of a cheating spouse. As I may have tried, none of the infidelity claims (with proof AND admission) was taken into account.
Good man. Lovely place. Spoilt by some of its inhabitants, but only some:-).
I see nothing wrong with the inhabitants. I can't fucking understand them half the time, but when I do, I can't stop laughing long enough to breathe.
What scheme could possibly be so general?
All gas/grocery purchases go on the HSBC Platinum Cash card, which may no longer be available. Earns 5% cash back at those two merchant types.
All restaurant purchases go on Costco American Express card, which is available in the US, anyhow. Earns 3% cash back at restaurants.
All other purchases go on Charles Schwab Invest First card, which also may no longer be available. Earns 2% cash back on all purchases.
Admittedly, it's a little OCD, but when you add it up, you get to about $1200 per year. I do consider that to be real money, and I find it to be worth a bit of neurotic behavior.
If your family's outgoings are only around $1000/month, you're atypically Scottish. This likely contradicts the implied assumption that $12/year means nothing to you (but if it really does, would you like my Paypal address so you can set up a regular payment to me?);
I've been called a lot of things before, but never "Scottish". You've shown a great deal of prescience, however, as Scotland is my favorite part of the UK.
At any rate, that $1000 figure was pulled out of thin air as a bit of a round number for illustration. You should not assume that it relates in any way to my monthly cash outflows.
But if you're dealing in small amounts and we're still considering risk free money storage, have you considered (your local equivalent of) a cash Individual Savings Account? Tax-free, and interest rates over 2.5%;
Well, I'm not British, so no.
Refresh my memory. Aren't ISAs like our (US) Individual Retirement Accounts where you can contribute a little each year and then you get your savings tax-free when you retire? Not sure why you're recommending this.
What is more, you potentially miss out on cashback, i.e. proportion of the discount charged to the retailer channeled into your pocket.
I do not miss this. I get 2% cash back for most purchases. 3% cash back in restaurants, and 5% cash back at gas stations and grocery stores.
Kinda makes the float seem negligible, doesn't it?
In the UK, the card issuing bank is jointly liable (Consumer Credit Act)
Good to know. I don't know anything about UK law, so I'll just take your word for it.
Stop using your credit card as a credit line, and start using it as a way to get up to 56 days extra interest on your money
Have you looked at interest rates lately? The national average Money Market interest rate is 0.78%
Since you seem like a smart guy, I'll assume you can do better than the national average and can get 1% (probably about right, since I know of a better MMA yield, however, with limited withdrawals per month, you probably won't be able to get the full yield). I'm not willing to give you credit for 56 days of float, since you'll be making purchases throughout the month. I'll give you the 25 day grace period plus 1/2 month to get an average float of 365/12/2+25=40 days of float.
Ok, so per $1000 that you charge, guess what your daily interest earned is. $0.027 (2.7 cents) per $1000 per day, minus taxes (interest income is taxable). So if you charge a $100 grocery bill, your float is 0.0027*40=$0.108 - taxes. If you charge $1000, you could maybe earn a whole dollar - taxes! WOOT W00T!!!!111
Or something like that.
plus (often statutory) protections on purchases.
Like what? You have statutory protection against billing errors, but purchase protection? There's a little, but it doesn't amount to much, and there are requirements you have to meet to even qualify.
But then we are talking about different solutions.
Yes. I think I've made this point about 12 times already. Postgres is not able to compete with Oracle here because they are different solutions.
The feature added to Postgresql is a shared nothing replication. A shared everything(It is just storage which is shared right?) will be able to handle synchronization much better.
Yes, it is only storage that is shared. FWIW, I didn't invent the term "shared-everything".
But don't the RAC solution leave you with the storage system as a single point of failure then?
Yes, which means any HA RAC architecture relies upon HA storage. If your SAN goes down or offline, your RAC is going to crash with 100% probability.
In practice, I have to give mixed reviews to Oracle RAC. My background is that I own an enterprise architecture consulting firm. I know enough about a variety of technologies to produce an architecture that satisfies requirements, but if you have specific questions about RAC implementation, I'm not going to be able to supply you with any details. I've never implemented one myself.
So back to my experience in the real world. I've found that RAC's "single point of failure" is often inexperienced DBAs. RAC is extremely flexible, and requires an expert to set up and tune it, properly. I don't see many storage failures, as most of my clients just write a big, fat check to EMC and say, "Make sure it don't die."
So real world, I've seen some RACs that run for years without a downtime. Oracle patches are applied to the running cluster via rolling restart.
On the other hand, I've seen some RACs experience more downtime than I would expect with just a single Oracle instance on a single node. The reason? Inexperienced DBAs. Although I had one client who must have had the worst EMC SAN ever, because that thing crashed about once every 6 months, taking down their beautiful RAC with it.
So, yes. There is a single point of failure on the storage. Typically, HA storage is not difficult to achieve, so this is not normally an issue. It certainly can be, however.
Does Oracle RAC really always ensure that if a query B is issued after* an other transaction A is comitted, then B will always see the result of A? Even if A and B are issued from different clients, and issued to different database servers?
Yes, it does.
Oracle RAC uses a "shared-everything" design where multiple Oracle instances access a single database. It is not possible for one RAC node to "fall behind" another node.
And those queries return incorrect results when they are behind the master. Pretty cool, huh?
Yup, that's the idea behind "asynchronous replication", and that's exactly how it's billed. If that's a deal-killer for you (for a large number of read-only queries it won't be.. think web applications where an eventually consistent state is perfectly fine)
Oh, believe me, I understand this very well. The poster I was replying to claimed that Postgres's new replication solution was "an awesome feature which lets Postgres compete with the big boys like Oracle."
My response then, and continues to be, that while a decent replication solution in Postgres core is an extremely welcome addition; it does not, in any way, shape, or form, compete with Oracle RAC. It's just not in the same league.
Don't get me wrong, Postgres is great, especially considering its price. It meets the requirements of most RDBMS use cases. Most applications do not need Oracle RAC. But if you do, then Postgres is not an option for you.
Not to mention you'll find plenty of MySQL DBAs with large multimaster deployments who swear calling it "replication", let alone multimaster replication generous even when everything is working correctly.
Why, exactly, are they running a large multimaster deployment? Instead of Cluster NDB, anyhow?
Running a large MySQL multimaster "cluster" sounds painful. I'm not a MySQL expert, but my understanding is that MySQL replication was not intended to handle that.
Wait, is that really a released, supported, and ready for production version? I thought mysql was still working on getting 6.x out the door. Cool if it is.
FYI, the version # of MySQL Cluster does not track with the version of the DBMS.
I'm really surprised there aren't more comments about SR/HS, as it's an awesome feature which lets Postgres compete with the big boys like Oracle.
Um, no.
Imagine having your expensive database server be dedicated *only* to writes, and having all your read-only queries spread across one or more slave(s) which are also your backup servers. Pretty cool, huh?
And those queries return incorrect results when they are behind the master. Pretty cool, huh?
This is why Postgres/MySQL replication is not even remotely the same thing as Oracle RAC.
The article indicates the driver was going about 15mph over the speed limit. I'd say that was relatively easy to identify by sight.
I think you'll find that if you try your hand at speed estimation, you'll find your accuracy will be abysmal.
I don't care how long you work at it, or how many classes you take. Humans aren't evolved to estimate the rate of speed of fast-moving objects. Hell, an object that is far enough away will look to us like it's not moving at all.
Slashdot Mirror
So I wouldn't be too smug about the high-end users, because they are paying *more* for Android than they would pay for iPhone.
I pay under $20/mo for my Palm Pre. It would be under $30/mo for an Evo (have to include the 4G tax). That's with actual unlimited 4G data (limited only by your throughput, not 5GB), unlimited text, unlimited GPS, mobile-to-mobile calling (on any network), and whatever else they include that I don't care about.
I don't know how much an iphone costs, but I bet it's a lot fucking more than that. For me, switching to iPhone would be prohibitively expensive. It's a great device, but your argument that it's cheap is bogus.
The reason I didn't consider an upgrade to the Evo is that the device is simply too big.
I'm not really invested in one platform over another. If Sprint released a CDMA iPhone, I'd consider it. Not sure how I feel about it lacking a keyboard, but I'd give it a shot. And if it doesn't sync with google like WebOS and Android, then it's a nonstarter for me. All my shit's in google right now.
Ramble, ramble, ramble.
I'm sorry. Did the Democrats repeal the PATRIOT Act while I wasn't looking? Did they close Guantanamo Bay? Warrantless wiretapping?
Please tell me when I got my civil liberties back, because it still feels like they're missing.
I'm currently evaluating datacenters to host a new application. Burst.net was on the short list, but they aren't any longer.
-- I mod Bruce Perens (3872) funny, just to keep him wondering.
Damnit, I really wanted to mod you funny, just to keep you wondering, but I already commented in this article.
Curses!
Good thing we sent some fresh, new faces to Washington to enact Change We Can Believe In(TM). Things are totally different now.
Oh, wait....
I try not to ask "what is this?" or "define this" type of questions, but I try to ask problem solving questions to see their problem solving skills.
Yeah, that's why I like to do both. But I don't typically ask "define this", per se. I'll ask "What's the difference between?" instead. That way, if they don't have a solid understanding of something on their resume, I'll find out immediately.
Example candidate whom I did not hire had DOM and SAX on his resume: ... ... ... ... uh, yeah.
Me: What is the difference between DOM and SAX?
Candidate: They are both for Java XML programming.
Me: Yes, but they each approach that task in a different way. Can you please elaborate on that?
Candidate:
Me: Would you like me to do you the favor of crossing those buzzwords off your resume so we can move on?
Candidate:
Needless to say, I did not select that candidate.
Free advice to all your job seekers out there: Don't put anything on your resume unless you are prepared to discuss that single topic for the entire length of the interview. It probably won't happen, but if that one buzzword is what I'm hiring you for, you're going to look like an idiot if the best you can do is expand the acronym!
The whole idea of LAMP is that it's an easy-to-learn, easy-to-deploy stack. Any competent developer should be able to learn this, quickly. Even if you could assign a "LAMP Competence Metric" (say a 0-10 scale) to a person, a competent developer who is a 2 in LAMP will be a 9 much more quickly than an incompetent developer who is currently a 6.
When I hire coders, I like to see how quickly they can understand a system via standard UML architecture diagrams. I like to see if they can implement a basic logic task using their choice of language, and I want that implementation to be straightforward, and I want it to use the coder's chosen language's foundation libraries instead of reinventing the wheel.
I also like to get a feel for how experienced they are in working with any of the various system development processes. (Hint: "What's that?" is a bad answer. Cowboy Coders are unwelcome at my company.)
Regarding testing basic computer science concepts, I like to ask candidates to differentiate one thing from another. For example, given your list of concepts ("BNF, data normalization, OOP, MVC"), I'd ask:
But apart from that you really, really want to look at what causes connection problems - Apple is likely far less guilty than you think.
I'm not an RF engineer by any stretch of the imagination, but I'll offer you this anecdote:
My house is in a little bit of a tricky zone, from a cellular reception perspective. Verizon is the only carrier that gets perfect reception; Sprint gets reasonable reception, but can roam on Verizon; AT&T gets weak, but usable signal; T-Mo gets nothing.
Of my friends on AT&T, the ones with blackberries can use their phones just fine at my house (T-Mo blackberries do not work, for some reason). On the other hand, iPhones (3G and 3Gs) just do not work. At all. Not inside. Not outside. Not on the roof. Not in a tree. Not with green eggs and ham.
If you were to stop by my house, your iPhone would be reduced to a bulky iPod Touch.
I'm sure there's some logical explanation from an RF perspective, but at least now you know why I tend to blame Apple first. :P
(Sent from my iPhone.)
I didn't think an iPhone could hold a network connection long enough to post an entire comment.
This isn't really about cheating, and in most states, cheating isn't relevant in a divorce settlement. Welcome to the wonderful world of no-fault divorce.
Where Facebook can really cause problems is in custody battles, or in lying to a judge. Consider how this might work.
gavron: No, your honor, I don't have a drinking problem. I'll enjoy a glass of wine with a meal from time to time, but I don't have a drinking problem.
Meanwhile, on FB: gavron is so fucking hungover right now and doesn't remember a goddamn thing from last night. It's just as well, as he really doesn't want to know if the new urine stain on his carpet was created by him or his miniature schnauzer.
gavron: I can't afford such a high child support payment. The economy is so bad, I was forced to accept a job delivering pizzas. I can't find work in my field, despite trying as best I can.
Meanwhile, on FB: gavron just got a great new job at Initech as a middle manager. He is so excited that he doesn't have to deliver any more goddamn pizzas!
Even if you are not so stupid as to hang yourself like that on facebook, you can't be sure that your friends will be equally clever. ([gavon's drinking buddy] wants to congratulate gavon on his new job. Don't forget those TPS reports, man! Happy hour tonight?)
The problem is that divorce cases can get complicated, and it's very easy to let something slip. It's one thing to say the wrong thing; it's quite another to post the wrong thing publicly.
In general, infidelity is a non-issue in divorce cases. I've been divorced twice, both times because of a cheating spouse. As I may have tried, none of the infidelity claims (with proof AND admission) was taken into account.
Welcome to the magical world of no-fault divorce.
Good man. Lovely place. Spoilt by some of its inhabitants, but only some :-).
I see nothing wrong with the inhabitants. I can't fucking understand them half the time, but when I do, I can't stop laughing long enough to breathe.
What scheme could possibly be so general?
All gas/grocery purchases go on the HSBC Platinum Cash card, which may no longer be available. Earns 5% cash back at those two merchant types.
All restaurant purchases go on Costco American Express card, which is available in the US, anyhow. Earns 3% cash back at restaurants.
All other purchases go on Charles Schwab Invest First card, which also may no longer be available. Earns 2% cash back on all purchases.
Admittedly, it's a little OCD, but when you add it up, you get to about $1200 per year. I do consider that to be real money, and I find it to be worth a bit of neurotic behavior.
If your family's outgoings are only around $1000/month, you're atypically Scottish. This likely contradicts the implied assumption that $12/year means nothing to you (but if it really does, would you like my Paypal address so you can set up a regular payment to me?);
I've been called a lot of things before, but never "Scottish". You've shown a great deal of prescience, however, as Scotland is my favorite part of the UK.
At any rate, that $1000 figure was pulled out of thin air as a bit of a round number for illustration. You should not assume that it relates in any way to my monthly cash outflows.
But if you're dealing in small amounts and we're still considering risk free money storage, have you considered (your local equivalent of) a cash Individual Savings Account? Tax-free, and interest rates over 2.5%;
Well, I'm not British, so no.
Refresh my memory. Aren't ISAs like our (US) Individual Retirement Accounts where you can contribute a little each year and then you get your savings tax-free when you retire? Not sure why you're recommending this.
What is more, you potentially miss out on cashback, i.e. proportion of the discount charged to the retailer channeled into your pocket.
I do not miss this. I get 2% cash back for most purchases. 3% cash back in restaurants, and 5% cash back at gas stations and grocery stores.
Kinda makes the float seem negligible, doesn't it?
In the UK, the card issuing bank is jointly liable (Consumer Credit Act)
Good to know. I don't know anything about UK law, so I'll just take your word for it.
In the US, this does not apply.
Stop using your credit card as a credit line, and start using it as a way to get up to 56 days extra interest on your money
Have you looked at interest rates lately? The national average Money Market interest rate is 0.78%
Since you seem like a smart guy, I'll assume you can do better than the national average and can get 1% (probably about right, since I know of a better MMA yield, however, with limited withdrawals per month, you probably won't be able to get the full yield). I'm not willing to give you credit for 56 days of float, since you'll be making purchases throughout the month. I'll give you the 25 day grace period plus 1/2 month to get an average float of 365/12/2+25=40 days of float.
Ok, so per $1000 that you charge, guess what your daily interest earned is. $0.027 (2.7 cents) per $1000 per day, minus taxes (interest income is taxable). So if you charge a $100 grocery bill, your float is 0.0027*40=$0.108 - taxes. If you charge $1000, you could maybe earn a whole dollar - taxes! WOOT W00T!!!!111
Or something like that.
plus (often statutory) protections on purchases.
Like what? You have statutory protection against billing errors, but purchase protection? There's a little, but it doesn't amount to much, and there are requirements you have to meet to even qualify.
But then we are talking about different solutions.
Yes. I think I've made this point about 12 times already. Postgres is not able to compete with Oracle here because they are different solutions.
The feature added to Postgresql is a shared nothing replication. A shared everything(It is just storage which is shared right?) will be able to handle synchronization much better.
Yes, it is only storage that is shared. FWIW, I didn't invent the term "shared-everything".
But don't the RAC solution leave you with the storage system as a single point of failure then?
Yes, which means any HA RAC architecture relies upon HA storage. If your SAN goes down or offline, your RAC is going to crash with 100% probability.
In practice, I have to give mixed reviews to Oracle RAC. My background is that I own an enterprise architecture consulting firm. I know enough about a variety of technologies to produce an architecture that satisfies requirements, but if you have specific questions about RAC implementation, I'm not going to be able to supply you with any details. I've never implemented one myself.
So back to my experience in the real world. I've found that RAC's "single point of failure" is often inexperienced DBAs. RAC is extremely flexible, and requires an expert to set up and tune it, properly. I don't see many storage failures, as most of my clients just write a big, fat check to EMC and say, "Make sure it don't die."
So real world, I've seen some RACs that run for years without a downtime. Oracle patches are applied to the running cluster via rolling restart.
On the other hand, I've seen some RACs experience more downtime than I would expect with just a single Oracle instance on a single node. The reason? Inexperienced DBAs. Although I had one client who must have had the worst EMC SAN ever, because that thing crashed about once every 6 months, taking down their beautiful RAC with it.
So, yes. There is a single point of failure on the storage. Typically, HA storage is not difficult to achieve, so this is not normally an issue. It certainly can be, however.
Does Oracle RAC really always ensure that if a query B is issued after* an other transaction A is comitted, then B will always see the result of A? Even if A and B are issued from different clients, and issued to different database servers?
Yes, it does.
Oracle RAC uses a "shared-everything" design where multiple Oracle instances access a single database. It is not possible for one RAC node to "fall behind" another node.
I think your MySQL spec sheet is about 10 years out of date.
And those queries return incorrect results when they are behind the master. Pretty cool, huh?
Yup, that's the idea behind "asynchronous replication", and that's exactly how it's billed. If that's a deal-killer for you (for a large number of read-only queries it won't be.. think web applications where an eventually consistent state is perfectly fine)
Oh, believe me, I understand this very well. The poster I was replying to claimed that Postgres's new replication solution was "an awesome feature which lets Postgres compete with the big boys like Oracle."
My response then, and continues to be, that while a decent replication solution in Postgres core is an extremely welcome addition; it does not, in any way, shape, or form, compete with Oracle RAC. It's just not in the same league.
Don't get me wrong, Postgres is great, especially considering its price. It meets the requirements of most RDBMS use cases. Most applications do not need Oracle RAC. But if you do, then Postgres is not an option for you.
Not to mention you'll find plenty of MySQL DBAs with large multimaster deployments who swear calling it "replication", let alone multimaster replication generous even when everything is working correctly.
Why, exactly, are they running a large multimaster deployment? Instead of Cluster NDB, anyhow?
Running a large MySQL multimaster "cluster" sounds painful. I'm not a MySQL expert, but my understanding is that MySQL replication was not intended to handle that.
Wait, is that really a released, supported, and ready for production version? I thought mysql was still working on getting 6.x out the door. Cool if it is.
FYI, the version # of MySQL Cluster does not track with the version of the DBMS.
You're representing slony as anything other than a fragile, unscalable, dirty hack? Wow, just wow.
I'll be interested to see how native replication performs. Including it in core was long-overdue, and I'm glad they finally got it in there.
I'm really surprised there aren't more comments about SR/HS, as it's an awesome feature which lets Postgres compete with the big boys like Oracle.
Um, no.
Imagine having your expensive database server be dedicated *only* to writes, and having all your read-only queries spread across one or more slave(s) which are also your backup servers. Pretty cool, huh?
And those queries return incorrect results when they are behind the master. Pretty cool, huh?
This is why Postgres/MySQL replication is not even remotely the same thing as Oracle RAC.
What does an HTTP server have to do with a DB?
PostgreSQL has a lot of more enterprisy features than MySQL.
You mean like clustering?
Oh, wait...
I don't understand. What is the advantage of PostgreSQL verses MySQL or a seperate HTTP server?
Forget it. I know someone is going to freak out and mod me troll. I do not intend to cause harm.
For your sake, I hope that was a troll.
By the way, what's the advantage of a car versus a truck, or a separate gas station?
The article indicates the driver was going about 15mph over the speed limit. I'd say that was relatively easy to identify by sight.
I think you'll find that if you try your hand at speed estimation, you'll find your accuracy will be abysmal.
I don't care how long you work at it, or how many classes you take. Humans aren't evolved to estimate the rate of speed of fast-moving objects. Hell, an object that is far enough away will look to us like it's not moving at all.
This ruling stinks.