Depends on how long you want to persist that data. If all you need to do is display, what is the benefit from loading it into a class rather than just accessing fields in the recordset? Even if you do need to persist, what benefit do you gain from caching the object as opposed to caching the recordset? Keep in mind the possibility of using client-side cursors or disconnected recordsets.
If you need to perform updates, putting your business logic in an object, rather than a stored procedure, can have its own problems. If you don't use a stored procedure, you compromise data integrity and security at the database level. If you do, why do you need the object? If the operation is part of a transaction that includes calls to other systems, you might be able to make a case. Although, I would prefer to provide the stored procedure with a means of making that call.
Tech support for your family??
on
Family Tech Support
·
· Score: 5, Insightful
Short answer: Don't do it.
Long Answer: Don't do it. It isn't worth the aggravation. When something goes wrong, it's automatically your fault. It doesn't matter they dropped the box while they were moving and unseated the boards. It's still your fault. It doesn't matter that they tested the huge electro- magnet for the science fair project right next to the hard drive. They still expect you to fix it over the phone.
If they can't put it together themselves after you tell them what parts to get and install an OS on their own, just let them buy the Dell and deal with their tech support department.
You get your data out by any application that knows how to use that object or its parents, in a classed OO language - which doesn't have to be the same application, if you design decently reusable classes
You missed the forest for the trees. If you store your data in the database, you can get it out and work with it from any language or application that can in some way communicate with the database. ODBC anyone?
Lets say that I need to generate an output file that will be sent to one of our associates datacenter. Lets say the file has 15 million records. Guess what? I'm not going to use an object to individually extract all the data for all 15 million rows.
It seems to me prima facie that good database design includes modeling the business reality & logic accurately and getting the database into a reasonably high normal form - our prof says at least 3NF, maybe your line is higher.
You're on the right track. Your professor is mostly wrong. In the real world, performance considerations often require denormalizing your database. One of the in-house applications here started out highly normalized. It's primary purpose is reporting but there is a certain amount of transaction processing. The performance hit from 18 joins on the primary view prompted us to largely denormalize it.
Databases generally don't scale horizontally. You can't just add a couple more servers and magically increase your performance. For real world business applications, if your database isn't your bottleneck, you aren't getting the most bang for your buck. When your database is your bottleneck, your database needs to be designed for optimal performance on the most common tasks not what properties an object exposes.
I think you missed the point. People coding against the widgets shouldn't worry about the database structure. People coding the widgets absolutely have to know about the database structure. Database comes first because the design has to be optimized based on the type of activity. A transaction processing database tends to be structured differently than a reporting database. Basing your design on your widget leads to horribly designed databases. Basing your design on how it will be used leads to well designed databases.
Besides, storing your object in the database rather than storing your data in the database is probably the most retarded suggestion I've ever heard of. Outside of the specific application that knows how to work with the object, how do you get your data? Not to mention doing any type of reporting or analysis out of your database.
No, you are absolutely wrong about this. Before the government forced the baby bells to share lines with competitors, there effectively was no broadband.
Yeah, there was ISDN. Last time I looked at ISDN, the phone company wanted $800 installation fee and $300/mo for service. For a 128k connection. Plus the installation fee didn't the include the uninterruptable power supply required to make a phone call if the electric went out. Right.....
Cable? Cable companies generally don't seem to offer internet in a particular area until after someone starts to offer DSL. It doesn't make sense from their perspective. Let's see, offer cable internet to an unknown, possibly very small, number of subscribers for $40/mo or add 30 more porn pay per view channels at $4.95 for 90 minutes. If they see the DSL market has enough draw, they might consider offering the service.
Satellite. That works really well if you live in an apartment or have lots of trees around that you don't have permission to top or take down.
You know why forcing the bells to share lowered the cost of DSL? Becuase the bells were forced to share the lines at cost. Let me repeat that. At cost. In other words the bells aren't losing money. They may be losing sales if they charge a substantially higher fee for service over and above the cost of the line. That's their problem. They need to work out a more efficient process and reduce service costs.
As for your claim that if existing broadband was more expensive, a better, cheaper solution would be invented, it's spurious. Better, cheaper solutions for broadband have been constantly developed for years. None of them got implemented because it would reduce the margin on business class T & DS lines. It took the government stepping in before broadband took off.
I rather suspect from your arguments that you are an adherent of Rand. She has some interesting points but she fails to take into account that the basis of civilization is cooperation and possibly sacrifice for the greater good. When everyone acts in a truly selfish manner, civilization breaks down.
Infrastructure: The basic facilities, services, and installations needed for the functioning of a community or society.
Government has an obligation to regulate infrastructure. Infrastructure is necessary for the functioning of a society. The role of government is to keep society functioning. See a connection?
Another factor you've completely ignored is the cost to enter a market. Cost to enter an infrastructure market is astronomical. The first company into the market is virtually guaranteed a de facto monopoly. Do you honestly believe any company is going to spend the money to run their own residential lines and then try to compete with the Bells? Aside the fact that the Bells have already recouped their initial investment many times over while the newcomer will be operating in the red for years, the Bells will simply lower their prices for a year or two until the cash strapped upstart fails.
And finally, if something is infrastructure, necessary to the functioning of society, then it is something that people in rural areas are entitled to receive, even if it isn't highly profitable. And that is where government has to step in. If it is only slightly profitable to operate in a rural area, it will take far too long to recoup the initial investment to be considered viable in today's short term profit driven economy.
Now in this specific case, I do not consider DSL infrastructure. However, the phone lines are. Further, the Bells have repeatedly demonstrated their commitment to ensuring that new technology won't interfere with their profit margins on existing services. For example, Verizon does not offer DSL in my area. There are several local companies that tried to offer the service but failed because of deliberate obstruction by Verizon.
Forcing a monopoly to allow competition is theft, how?
1) Powerpoint. Executives love presentations. Animated graphics. Sound clips. Scripting. Embed documents, spreadsheets, pictures, just about anything. The flashier the better.
2) Just about everyone has a windows PC at home. They know how to turn it on and log in. They know how to get to email, web browser, word processor, etc. In other words, the company doesn't have to pay for training for most employees to be able to use the computer well enough to do their job. They still have to train them in particular applications, but not the basics.
3) Powerpoint.
4) X-terminals eat a lot of network bandwidth. We have around 20k customer service workstations. Any savings in desktop hardware and administration would be taken up by the increase in network hardware and administration.
5) If you aren't using X for terminals, and it sort of matters how stable your GUI app actually is, X might not be the best choice.
6) Powerpoint.
7) If most workers only need email and word processing, what are they doing the rest of their work on? You think our accountants still use 10 column green pads and write entries by hand into a bound ledger? How about our customer service staff? Think they walk back and pull the customer's folder out of a filing cabinet? How about the loan officers? Or the credit analysts? Or even the people that call directory assistance to make sure the phone number you put on your credit card application matches the mailing address? All of these jobs use custom applications with complex business logic. Workstations are cheap compared to beefier servers.
8) Microsoft makes some pretty decent back office products. They aren't as robust as products from other vendors. They also cost a lot less. Go compare SQL Server licensing to Oracle. If you don't need Oracle to do the job, why pay the extra? Sure, Postgres or MySQL are free. Until you want support.
9) Did I mention Powerpoint?
10) Big Business uses Windows desktops because it is ultimately cheaper for them to do so. It's all about the bottom line. Do you honestly think for a minute that if rolling out *nix desktops across the enterprise would generate a substantial savings, that it wouldn't be done?
*nix has it's place. We use it for our larger web farms, Oracle, DB2, etc. Mission critical applications. NT servers are used for smaller departmental web sites, databases too big for Access but not requiring Oracle, etc. Non mission critical applications. Everyone from the lowliest data entry operator up to the CEO uses a Windows desktop.
Case lock is fine and dandy on a desktop. On a server, access to the cd-rom to boot the recovery console requires physical access to the server room and physical access to the rack in which the server is located. Now, how does the case lock prevent someone from pulling the hot plug drives?
As for your comment on prevalent philosophy, I think you're oversimplifying. In my experience, it's more like:
"The anonymous access vulnerability to the sql/query pipe is not relevant. The effort required to successfully hijack an administrators connection from the pool is substantial and requires an existing user account. Additionally, disallowing anonymous access will probably require redesign and coding of existing applications so that they do not fail. The root exploit in SNMP on the other hand simply requires network access to the box. It is not a productive use of resources to fix anonymous access to the pipe at this time."
It's kind of like welding bars over your windows and leaving the front door unlocked. Or to put it another way, if removing the vulnerability incurs substantial cost while not significantly increasing security, it's a waste of time and money unless you are also going to fix the other vulnerabilities. In the corporate world, you don't always have the option of fixing some of these holes.
For example, I'm not concerned in the slightest about known vulnerabilities that would actually take some work to exploit on my servers. Not while corporate mandates that I have a remote shell service running as domain admin and doesn't even require a password to connect. Anyone that can find the port can telnet in and own my boxes. I'm supposed to be worried about exploits that actually require some coding? Doesn't mean I don't patch. It just means I don't patch if it's going to break something else important.
I work for a megacorp so my viewpoint may not apply to small business coding.
If they consistently don't do what they are told to do, why are they still on the project? Why are they still employed?
It costs the company significant time and resources to go back bring this type of code up to standard so it can be included in a production release. If other functionality that depends on this code has to be pulled from a scheduled release, it can cost millions.
Millions of dollars vs. a severence package for the guy who can't follow the rules. You do the math.
You're right. The guy hasn't paid his dues. He hasn't spent days poring over man pages. He hasn't spent hours trying to recompile his kernel to get it to recognize his NIC. He probably doesn't even know how to use xf86config.
Then again, maybe the point was you shouldn't have to be a wizard to get an office suite to talk to an odbc datasource. Maybe the point was that real people trying to do real work don't want to be a sysadmin. They just want to get their work done.
Lets burn some more Karma since replying to a post is apparently offtopic around here.
I didn't ask how using Apple was going to help my kid learn about computers. I didn't ask how OS X would make my kid a better person. I don't even recall mentioning Windows at all.
The question is "How does buying more expensive hardware that isn't generally used in the real world help my kid get a better education?".
My kid doesn't want to be a Comp Sci major. He doesn't want to learn computing concepts. He doesn't care which OS is better. He doesn't care if there are viable alternatives to Windows. He wants to use his computer to help him do better in school. He wants to use his computer to do research. Write papers. Create presentations.
So how does an Apple do this better than an Intel/AMD computer? For the reading comprehension impaired, let me point out once again I did not ask why is OS X better than Windows.
The point of "not generally used in the real world" is that this isn't even a mitigating factor that would help the case for Apple.
And finally, as a taxpayer and a parent who has to make sure that my son can do his homework on the same platform used in school, I see some distinct advantages to using the Intel/AMD platform. It doesnt put nearly as much of a dent in my wallet. The computer might be useful for something other than doing his homework. I can give my son my old dev rig whenever I upgrade. My taxes don't go up just so that my son can see that there are viable alternatives to Windows.
I'm not saying Apple is bad. I'm saying answer the same question I asked. I'm saying back up your assertions with some facts that have relevance to the general public. I'm saying that if Apple wants to compete in the educational market, they have provide a compelling price point.
He didn't say the database crashed. He said the database locked up. When someone tells me their database locked up because they are doing complex query on 10 million rows, and Oracle tech support tells them to take a hike, this suggests several possibilities to me.
1) He's running this on a box that can't handle the load. 2) He can't write a decent SQL statement. 3) He doesn't understand how to optimize. 4) He isn't qualified to do database design and needs to hire a consultant. 5) Any or All of the above.
We run terabyte data warehouses on Oracle. Billions of rows. We don't have issues like this guy. We run OLTP databases with 20k simultaneous users on Oracle. We don't have issues like this guy. Think about it. If Oracle can't handle 10 million rows, large corporations wouldn't use it.
If he isn't the developer, why is he allowed to run an arbitrary query against an Oracle database? If it's something he's got on his desktop, fine. But you never, ever, ever let some smuck run an arbitrary query against your production database. Because no matter how badly it's written, no matter how much cpu time it hogs from other applications, no matter how much it impacts all of the other users, the database does exactly what the query told it to do. If you need to provide some adhoc functionality that also protects the database from bad queries, that's where middleware enters the picture.
I've used my apparently already obsolete regular firewire port exactly zero times. So Firewire 800 will let me transfer nothing how many times faster? It's a stretch, but someone, somewhere might have a use for this. I'll give you this one.
Gigabit?. Sure. Just as soon as I plonk down another $500 for an 8 port gigabit switch to replace my $50 8 port 10/100 switch.
Wireless? Not included in $1999 price point. The antenna is included. Add $100 for the card. Add $250 for the base station. Did I mention that Amazon lists Linksys WMP54G PCI card at $70 and Linksys WAP54G access point at $130? And your workstation needs to be wireless why? You can't carry it around with you. With all of the normal cables what's one or two more?
Sweet Applications? You mean iTunes and iMovie? You have to upgrade to the superdrive for iDVD to work. They are nice applications but PC owners can get nice applications with all the money they saved and stil have some left over.
These features are like Picture in Picture on a TV. All they do is maintain a price point without adding any real value. You want to convince someone Apple is worth the extra money? You have to come up with a convincing cost/benefit analysis based on benefits they will actually use. Based on the actual benefit to me, Apple would have to cut the price on that $1999 model down to under $1000 w/o the superdrive or around $1200 with.
And it still doesn't even come close to the price/performance of an Intel or AMD box. Plus corporations use Intel based platforms because of that price/performance. So therefore corporations want to employ people who know how to use said Intel based platforms. Plus the higher cost of computing hardware means higher tutition for students or higher taxes for their parents or both.
Explain to me again how a school buying more expensive hardware that isn't generally used in the real world is going to help my kid get a better education?
Sure, you want to trick out a graphics arts lab at a University with some PowerMacs, great. But for general educational use, Apple is playing catch up and they need to slash prices to the bone if they want to compete.
Buying a video card with an eye to the future? Makes no sense. Product cycles are too short. Don't buy a $400 video card now because it does DX9 when DX9 games aren't available. By the time you actually get some use out of those features, you'll be able to buy a better card for a less money.
You are retrieving data from a 10 million row table using a very complex query and you are having performance problems? Who would have thought that?
Normally I get paid a lot of money to solve problems like this but I'll give you a little guidance for free since you didn't like Oracle's answer.
1) Maybe you should think about optimizing your query a bit. Running complex queries against 10+ rows can be problematic even when the RDMS has a good optimizer. Is there a less complex way to accomplish the same thing? If not, you may have to give the optimizer hints. Can you use an index to pull a smaller dataset into a working table where you do your complex operations?
2) Profile your system to determine where the bottleneck is to be found. Is it CPU bound or IO bound. If it's IO bound, would more memory help? Can your tablespace be spread across more disks? Would a beefier system be appropriate? Cost Effective?
This is why you hire qualified developers and administrators. I'm not surprised the tech team gave you that answer. You call the tech team when there is a real problem with the software. If you were paying Oracle to develop the database for you, you might have a case. But then, if that were true, you wouldn't have called tech support, would you?
This is disingenous. I have to reboot my fscking windows desktop 4 times a day at work. At work. Because the fscking PC support department just can't get it through their fscking thick skulls that they can't just push the same fscking image out to every PC. Because every batch of fscking PCs we order from the fscking vendor comes with different fscking internals even though they are the same fscking part number. Q:"Why is my PC broken again?" A:"I don't know. It worked in the lab."
On the other hand the NT servers I admin at work get rebooted when I have to patch. Period. Might be because I actually run the patch on each one instead of pushing some files and registry entries over and hoping I didn't miss anything. Then again maybe they just like me better.
At home I only reboot when some D3D game crashes my video drivers. That's hardly a Microsoft problem. It's a problem with the vendor supplied drivers.
Linux is stable. It's just irrelevant at the moment. I'd sooner go with OpenBSD on a server. As a desktop? It's a geek toy. People who just want to get things done don't use it. You know why? Because it takes longer to figure out how to get the hardware/software/user interface for a given task to work than it does to just do the work in another operating system.
This isn't directly related to the parent post but insomuch as the parent post seems to advocate Linux over Windows, allow me to relate my experience with Linux.
I've tried Linux twice. I tried it back around '97. Got the base system installed no problem. The 1024 cylinder boot limitation was annoying but not horrible. Configuring X was pretty horrible. Mostly because the kind of detailed specs X wanted about the monitor simply weren't available. PPP took me three days to get mostly working. I never did get it to consistently reconnect when the ISP did their hourly drop. X is probably the biggest reason I ended up dropping Linux. Call me a spoiled Windows user but I like being able to copy and or drag and drop without having to kill X from another box after it goes haywire. 3D gaming was pathetic. No real alternative to MS Office.
Tried it again a while ago with SUSE 7. Ran into the 1024 cylinder limitation. It was annoying on a 6GB drive. It was ridiculous years later on 20GB drive. Strike One. (Yes I know it's fixed now.) Trying to install any vaguely useful software quickly degenerated into darkest depths of conflicting libraries/ widget sets/ toolkits/ desktops/ etc. Dll Hell was never _this_ bad. Strike Two. (It isn't as bad now.) I had to recompile the kernal so it would recognize my NIC. Unfortunately something had apparently knackered gcc during strike two. (Probably fixed. I don't really know or care what the root cause was.) Strike Three.
I get paid decent money to admin computers at my job. I don't get paid to do the same at home. Time I spend trying to coerce and cojole Linux into doing the same tasks that other OS provide out of the box is less time with my family. If I'm not getting paid for it, and I wouldn't have to spend that time using a different OS, then it's a flat out waste of my time. You want Linux on the desktop? Get a version of Linux that provides the close to the same functionality of Windows or OS X out of the fscking box, and more people might take you seriously.
I really don't think the point is to boost enlistment rates. The point is raising the quality of the recruits they do get.
It takes years to turn a raw recruit into a soldier that can make good decisions with bullets flying and the plan in tatters. To do so strictly with field exercises is extremely time consuming and expensive. For small scale engagements with prep time and after action reviews, you're pushing it to get 6-10 run throughs in a day. For larger operations you might run through once or twice in a day. If a parachute drop is required, just plan on once. Period. Plus, brass isn't cheap and typically it doesn't get policed up except on a range.
A video game like America's Army allows training missions to be run at a much faster pace with little expense. It teaches cooperation, teamwork, and most importantly, tactical thinking. In other words it's a head start for the small unit leaders of tomorrow. As long as the game is realistic enough that the optimal tactics in the game carry over substantially in the real world, it's a good thing.
If ends up boosting the enlistment rate, so what? Is it a bad thing to serve your country? Is it wrong to defend your loved ones? If your objection is simply that it might make war seem cool, I can assure you that if basic training or assimilation into their unit doesn't rid them of that notion, either they aren't combat arms or they won't be for long.
It's interesting they include a PC in the tuition for the first term. $12k vs $7k for the second term. Implies a $4k-$5k price tag for said computer. Ownership doesn't transfer until after the fourth term (and a total expenditure of $29k). What a rip off.
Why not just post minimum specs each term? If a student can come up with tuition for this program, they can afford a few hundred each term to upgrade their PC if required.
Floppy Disk? By the time programs came on floppies, I was so sick of typing the programs that I was happy to pay the extra. Besides, I never did care for the noise the cassette tapes made.
Regardless of whether DRM can leverage TCPA, TCPA is an important step in the right direction for business computing. Its probably unnecessary for a consumer system. I say probably because I haven't thought of a good use other than DRM. I'm not going to make the assumption that because I haven't thought of a good use that there isn't one.
In business computing, the ability to restrict hardware and software will become a priority. For example, being able to ensure that account information was not compromised even if one of the thousands of employees with physical access to a customer service workstation installed additional software on the machine. Yes, it's against policy and they will be fired for it. Yes, we have the machines as locked down as they can be and still function. Yes, we do FBI background checks on every employee. And, yes, they can still write them down. But, if you have physical access to a computer, you can alter the operating system. If someone walks out with a file containing thousands of accounts it's a much bigger problem than a couple names and numbers scribbled on a napkin.
Aside from making it difficult to steal large amounts of sensitive information from the company, trusted hardware makes biometrics a viable form of authentication. Sure, biometrics has issues. So do static passwords. Most of the issues with biometrics can be solved through trusted hardware. The significant remaining issue, fooling the trusted hardware, can be mitigated through an additional means of authentication. Probably the smart card built into your employee ID. It isn't infallible but it's certainly more secure than passwords. Stealing an ID and forging a fingerprint or retina scan or DNA sample vs. social engineering a static password, which do you think is harder?
Frankly I don't give a damn if some software DRM application tries to leverage TCPA on my home machine. Call me old fashioned but I'm not about to pay for an ebook or an mp3 or a divx movie. What do you get? Some zeros and ones that go away if your hard drive fails? Nah..I'll pass. And if they want to stop me from ripping a CD to mp3s? Well then I can alter the hardware and/or use a different OS.
A key point on the issue of consent was raised in the comments to the original article. This or a similar case is probably why Dark Age of Camelot requires you to scroll through the EULA and rules of conduct before the Agree button can be clicked. I think this practice will become more widespread in the future.
Re: Cheap Trick: RealPlayer Windows Installation (Score: 1)
by Barry on Thursday, January 16 @ 11:43:36 EST (User Info | Send a Message) www.wbklaw.com The practive may or may not be "unconscionable," but it does likely fail to create an enforceable contract, at least in the 2nd Circuit. See below.
In Specht v. Netscape Communications Corp.,[*] the court stated that, to be enforceable, click-wrap and shrink-wrap agreements require users to perform an affirmative action unambiguously expressing assent. In this case, there was one small box of text referring to the license language, which was "couched in the mild request 'Please review and agree . ..'" The court found that this was a "mere invitation" and not a condition. Failure to require users to indicate assent was "fatal" to Netscape's argument that a contract had been formed. To be enforceable, language must indicate that a user "must" agree to the license terms. Importantly, for purposes of this post, on appeal, the @nd Circuit found it important that the notice of the existence of the contract was not visible until a user scrolled to the next screen and not visible when the user made the decision to "accept." The Second Circuit said that a reasonably prudent person would not have known or learned of the license terms prior to using the software and could not be held to have had "constructive notice" of the terms. The Second Circuit held that "in circumstances such as these, where consumers are urged to download free software at the immediate click of a button, a reference to the existence of license terms on a submerged screen is not sufficient to place consumers on inquiry or constructive notice of those terms." If the hidden terms at issue here ("consenting" to unwanted spam) would be material to the user's decision to click/accept, Specht may be instructive.
[*] Specht, 150 F.Supp.2d 585 (S.D. NY 2001), aff'd Specht, et al. v. Netscape Communications Corp., et al., Nos. 01-7860(L), 01-7870 (CON), 01-7872(CON) (2nd Cir. October 1, 2002).
The class of games may be well defined, but this tournament is not. How long do the humans have to study payoff matrix between each move? What kind of background will the humans come from? I expect a PhD in game theory is going to play differently from a housewife.
In a simple game like this, it's possible to write a program to emulate the behavior of a particular type of person. Getting the best match on an unspecified group of people involves a large element of blind luck.
Palladium will go through. Trusted Computing will happen. It doesn't matter whether its used for DRM or not. It will happen because it provides an increased level of security and accountability for business computing.
For example, biometric authentication is basically worthless at the moment. It's all too easy to spoof, vulnerable to replay attacks, vulnerable to hardware modifications, etc. If you set up a trusted system that only accepts known hardware and software, biometrics gets a lot closer to being a reality.
As an individual, I'm not happy about trusted computing because I see all the ways it can be abused.
As a developer/sysadmin, I'm looking forward to incorporating this into our applications. No more password reset calls. No more trying to remember 15 different strong passwords without writing them down. A big fat raise because my boss won't have to remember all his passwords without writing them down.
Slashdot Mirror
Depends on how long you want to persist that data. If all you need to do is display, what is the benefit from loading it into a class rather than just accessing fields in the recordset? Even if you do need to persist, what benefit do you gain from caching the object as opposed to caching the recordset? Keep in mind the possibility of using client-side cursors or disconnected recordsets.
If you need to perform updates, putting your business logic in an object, rather than a stored procedure, can have its own problems. If you don't use a stored procedure, you compromise data integrity and security at the database level. If you do, why do you need the object? If the operation is part of a transaction that includes calls to other systems, you might be able to make a case. Although, I would prefer to provide the stored procedure with a means of making that call.
Short answer: Don't do it.
Long Answer: Don't do it. It isn't worth the aggravation. When something goes wrong, it's automatically your fault. It doesn't matter they dropped the box while they were moving and unseated the boards. It's still your fault. It doesn't matter that they tested the huge electro- magnet for the science fair project right next to the hard drive. They still expect you to fix it over the phone.
If they can't put it together themselves after you tell them what parts to get and install an OS on their own, just let them buy the Dell and deal with their tech support department.
You get your data out by any application that knows how to use that object or its parents, in a classed OO language - which doesn't have to be the same application, if you design decently reusable classes
You missed the forest for the trees. If you store your data in the database, you can get it out and work with it from any language or application that can in some way communicate with the database. ODBC anyone?
Lets say that I need to generate an output file that will be sent to one of our associates datacenter. Lets say the file has 15 million records. Guess what? I'm not going to use an object to individually extract all the data for all 15 million rows.
It seems to me prima facie that good database design includes modeling the business reality & logic accurately and getting the database into a reasonably high normal form - our prof says at least 3NF, maybe your line is higher.
You're on the right track. Your professor is mostly wrong. In the real world, performance considerations often require denormalizing your database. One of the in-house applications here started out highly normalized. It's primary purpose is reporting but there is a certain amount of transaction processing. The performance hit from 18 joins on the primary view prompted us to largely denormalize it.
Databases generally don't scale horizontally. You can't just add a couple more servers and magically increase your performance. For real world business applications, if your database isn't your bottleneck, you aren't getting the most bang for your buck. When your database is your bottleneck, your database needs to be designed for optimal performance on the most common tasks not what properties an object exposes.
I think you missed the point. People coding against the widgets shouldn't worry about the database structure. People coding the widgets absolutely have to know about the database structure. Database comes first because the design has to be optimized based on the type of activity. A transaction processing database tends to be structured differently than a reporting database. Basing your design on your widget leads to horribly designed databases. Basing your design on how it will be used leads to well designed databases.
Besides, storing your object in the database rather than storing your data in the database is probably the most retarded suggestion I've ever heard of. Outside of the specific application that knows how to work with the object, how do you get your data? Not to mention doing any type of reporting or analysis out of your database.
No, you are absolutely wrong about this. Before the government forced the baby bells to share lines with competitors, there effectively was no broadband.
Yeah, there was ISDN. Last time I looked at ISDN, the phone company wanted $800 installation fee and $300/mo for service. For a 128k connection. Plus the installation fee didn't the include the uninterruptable power supply required to make a phone call if the electric went out. Right.....
Cable? Cable companies generally don't seem to offer internet in a particular area until after someone starts to offer DSL. It doesn't make sense from their perspective. Let's see, offer cable internet to an unknown, possibly very small, number of subscribers for $40/mo or add 30 more porn pay per view channels at $4.95 for 90 minutes. If they see the DSL market has enough draw, they might consider offering the service.
Satellite. That works really well if you live in an apartment or have lots of trees around that you don't have permission to top or take down.
You know why forcing the bells to share lowered the cost of DSL? Becuase the bells were forced to share the lines at cost. Let me repeat that. At cost. In other words the bells aren't losing money. They may be losing sales if they charge a substantially higher fee for service over and above the cost of the line. That's their problem. They need to work out a more efficient process and reduce service costs.
As for your claim that if existing broadband was more expensive, a better, cheaper solution would be invented, it's spurious. Better, cheaper solutions for broadband have been constantly developed for years. None of them got implemented because it would reduce the margin on business class T & DS lines. It took the government stepping in before broadband took off.
I rather suspect from your arguments that you are an adherent of Rand. She has some interesting points but she fails to take into account that the basis of civilization is cooperation and possibly sacrifice for the greater good. When everyone acts in a truly selfish manner, civilization breaks down.
Ok Troll, I'll bite.
Infrastructure: The basic facilities, services, and installations needed for the functioning of a community or society.
Government has an obligation to regulate infrastructure. Infrastructure is necessary for the functioning of a society. The role of government is to keep society functioning. See a connection?
Another factor you've completely ignored is the cost to enter a market. Cost to enter an infrastructure market is astronomical. The first company into the market is virtually guaranteed a de facto monopoly. Do you honestly believe any company is going to spend the money to run their own residential lines and then try to compete with the Bells? Aside the fact that the Bells have already recouped their initial investment many times over while the newcomer will be operating in the red for years, the Bells will simply lower their prices for a year or two until the cash strapped upstart fails.
And finally, if something is infrastructure, necessary to the functioning of society, then it is something that people in rural areas are entitled to receive, even if it isn't highly profitable. And that is where government has to step in. If it is only slightly profitable to operate in a rural area, it will take far too long to recoup the initial investment to be considered viable in today's short term profit driven economy.
Now in this specific case, I do not consider DSL infrastructure. However, the phone lines are. Further, the Bells have repeatedly demonstrated their commitment to ensuring that new technology won't interfere with their profit margins on existing services. For example, Verizon does not offer DSL in my area. There are several local companies that tried to offer the service but failed because of deliberate obstruction by Verizon.
Forcing a monopoly to allow competition is theft, how?
Why do corporations use Microsoft?
1) Powerpoint. Executives love presentations. Animated graphics. Sound clips. Scripting. Embed documents, spreadsheets, pictures, just about anything. The flashier the better.
2) Just about everyone has a windows PC at home. They know how to turn it on and log in. They know how to get to email, web browser, word processor, etc. In other words, the company doesn't have to pay for training for most employees to be able to use the computer well enough to do their job. They still have to train them in particular applications, but not the basics.
3) Powerpoint.
4) X-terminals eat a lot of network bandwidth. We have around 20k customer service workstations. Any savings in desktop hardware and administration would be taken up by the increase in network hardware and administration.
5) If you aren't using X for terminals, and it sort of matters how stable your GUI app actually is, X might not be the best choice.
6) Powerpoint.
7) If most workers only need email and word processing, what are they doing the rest of their work on? You think our accountants still use 10 column green pads and write entries by hand into a bound ledger? How about our customer service staff? Think they walk back and pull the customer's folder out of a filing cabinet? How about the loan officers? Or the credit analysts? Or even the people that call directory assistance to make sure the phone number you put on your credit card application matches the mailing address? All of these jobs use custom applications with complex business logic. Workstations are cheap compared to beefier servers.
8) Microsoft makes some pretty decent back office products. They aren't as robust as products from other vendors. They also cost a lot less. Go compare SQL Server licensing to Oracle. If you don't need Oracle to do the job, why pay the extra? Sure, Postgres or MySQL are free. Until you want support.
9) Did I mention Powerpoint?
10) Big Business uses Windows desktops because it is ultimately cheaper for them to do so. It's all about the bottom line. Do you honestly think for a minute that if rolling out *nix desktops across the enterprise would generate a substantial savings, that it wouldn't be done?
*nix has it's place. We use it for our larger web farms, Oracle, DB2, etc. Mission critical applications. NT servers are used for smaller departmental web sites, databases too big for Access but not requiring Oracle, etc. Non mission critical applications. Everyone from the lowliest data entry operator up to the CEO uses a Windows desktop.
Case lock is fine and dandy on a desktop. On a server, access to the cd-rom to boot the recovery console requires physical access to the server room and physical access to the rack in which the server is located. Now, how does the case lock prevent someone from pulling the hot plug drives?
As for your comment on prevalent philosophy, I think you're oversimplifying. In my experience, it's more like:
"The anonymous access vulnerability to the sql/query pipe is not relevant. The effort required to successfully hijack an administrators connection from the pool is substantial and requires an existing user account. Additionally, disallowing anonymous access will probably require redesign and coding of existing applications so that they do not fail. The root exploit in SNMP on the other hand simply requires network access to the box. It is not a productive use of resources to fix anonymous access to the pipe at this time."
It's kind of like welding bars over your windows and leaving the front door unlocked. Or to put it another way, if removing the vulnerability incurs substantial cost while not significantly increasing security, it's a waste of time and money unless you are also going to fix the other vulnerabilities. In the corporate world, you don't always have the option of fixing some of these holes.
For example, I'm not concerned in the slightest about known vulnerabilities that would actually take some work to exploit on my servers. Not while corporate mandates that I have a remote shell service running as domain admin and doesn't even require a password to connect. Anyone that can find the port can telnet in and own my boxes. I'm supposed to be worried about exploits that actually require some coding? Doesn't mean I don't patch. It just means I don't patch if it's going to break something else important.
I work for a megacorp so my viewpoint may not apply to small business coding.
If they consistently don't do what they are told to do, why are they still on the project? Why are they still employed?
It costs the company significant time and resources to go back bring this type of code up to standard so it can be included in a production release. If other functionality that depends on this code has to be pulled from a scheduled release, it can cost millions.
Millions of dollars vs. a severence package for the guy who can't follow the rules. You do the math.
You're right. The guy hasn't paid his dues. He hasn't spent days poring over man pages. He hasn't spent hours trying to recompile his kernel to get it to recognize his NIC. He probably doesn't even know how to use xf86config.
Then again, maybe the point was you shouldn't have to be a wizard to get an office suite to talk to an odbc datasource. Maybe the point was that real people trying to do real work don't want to be a sysadmin. They just want to get their work done.
Lets burn some more Karma since replying to a post is apparently offtopic around here.
I didn't ask how using Apple was going to help my kid learn about computers. I didn't ask how OS X would make my kid a better person. I don't even recall mentioning Windows at all.
The question is "How does buying more expensive hardware that isn't generally used in the real world help my kid get a better education?".
My kid doesn't want to be a Comp Sci major. He doesn't want to learn computing concepts. He doesn't care which OS is better. He doesn't care if there are viable alternatives to Windows. He wants to use his computer to help him do better in school. He wants to use his computer to do research. Write papers. Create presentations.
So how does an Apple do this better than an Intel/AMD computer? For the reading comprehension impaired, let me point out once again I did not ask why is OS X better than Windows.
The point of "not generally used in the real world" is that this isn't even a mitigating factor that would help the case for Apple.
And finally, as a taxpayer and a parent who has to make sure that my son can do his homework on the same platform used in school, I see some distinct advantages to using the Intel/AMD platform. It doesnt put nearly as much of a dent in my wallet. The computer might be useful for something other than doing his homework. I can give my son my old dev rig whenever I upgrade. My taxes don't go up just so that my son can see that there are viable alternatives to Windows.
I'm not saying Apple is bad. I'm saying answer the same question I asked. I'm saying back up your assertions with some facts that have relevance to the general public. I'm saying that if Apple wants to compete in the educational market, they have provide a compelling price point.
He didn't say the database crashed. He said the database locked up. When someone tells me their database locked up because they are doing complex query on 10 million rows, and Oracle tech support tells them to take a hike, this suggests several possibilities to me.
1) He's running this on a box that can't handle the load.
2) He can't write a decent SQL statement.
3) He doesn't understand how to optimize.
4) He isn't qualified to do database design and needs to hire a consultant.
5) Any or All of the above.
We run terabyte data warehouses on Oracle. Billions of rows. We don't have issues like this guy. We run OLTP databases with 20k simultaneous users on Oracle. We don't have issues like this guy. Think about it. If Oracle can't handle 10 million rows, large corporations wouldn't use it.
If he isn't the developer, why is he allowed to run an arbitrary query against an Oracle database? If it's something he's got on his desktop, fine. But you never, ever, ever let some smuck run an arbitrary query against your production database. Because no matter how badly it's written, no matter how much cpu time it hogs from other applications, no matter how much it impacts all of the other users, the database does exactly what the query told it to do. If you need to provide some adhoc functionality that also protects the database from bad queries, that's where middleware enters the picture.
Lets cover these in order.
I've used my apparently already obsolete regular firewire port exactly zero times. So Firewire 800 will let me transfer nothing how many times faster? It's a stretch, but someone, somewhere might have a use for this. I'll give you this one.
Gigabit?. Sure. Just as soon as I plonk down another $500 for an 8 port gigabit switch to replace my $50 8 port 10/100 switch.
Wireless? Not included in $1999 price point. The antenna is included. Add $100 for the card. Add $250 for the base station. Did I mention that Amazon lists Linksys WMP54G PCI card at $70 and Linksys WAP54G access point at $130? And your workstation needs to be wireless why? You can't carry it around with you. With all of the normal cables what's one or two more?
Sweet Applications? You mean iTunes and iMovie? You have to upgrade to the superdrive for iDVD to work. They are nice applications but PC owners can get nice applications with all the money they saved and stil have some left over.
These features are like Picture in Picture on a TV. All they do is maintain a price point without adding any real value. You want to convince someone Apple is worth the extra money? You have to come up with a convincing cost/benefit analysis based on benefits they will actually use. Based on the actual benefit to me, Apple would have to cut the price on that $1999 model down to under $1000 w/o the superdrive or around $1200 with.
And it still doesn't even come close to the price/performance of an Intel or AMD box. Plus corporations use Intel based platforms because of that price/performance. So therefore corporations want to employ people who know how to use said Intel based platforms. Plus the higher cost of computing hardware means higher tutition for students or higher taxes for their parents or both.
Explain to me again how a school buying more expensive hardware that isn't generally used in the real world is going to help my kid get a better education?
Sure, you want to trick out a graphics arts lab at a University with some PowerMacs, great. But for general educational use, Apple is playing catch up and they need to slash prices to the bone if they want to compete.
Buying a video card with an eye to the future? Makes no sense. Product cycles are too short. Don't buy a $400 video card now because it does DX9 when DX9 games aren't available. By the time you actually get some use out of those features, you'll be able to buy a better card for a less money.
Enterprise Manager
All Tasks->Generate SQL Scripts
Make sure you check the boxes to include constraints, foreign keys, indexes, etc.
You can even script the Agent jobs. Don't know if anything else can read them though.
Hope you aren't using DTS packages.
You are retrieving data from a 10 million row table using a very complex query and you are having performance problems? Who would have thought that?
Normally I get paid a lot of money to solve problems like this but I'll give you a little guidance for free since you didn't like Oracle's answer.
1) Maybe you should think about optimizing your query a bit. Running complex queries against 10+ rows can be problematic even when the RDMS has a good optimizer. Is there a less complex way to accomplish the same thing? If not, you may have to give the optimizer hints. Can you use an index to pull a smaller dataset into a working table where you do your complex operations?
2) Profile your system to determine where the bottleneck is to be found. Is it CPU bound or IO bound. If it's IO bound, would more memory help? Can your tablespace be spread across more disks? Would a beefier system be appropriate? Cost Effective?
This is why you hire qualified developers and administrators. I'm not surprised the tech team gave you that answer. You call the tech team when there is a real problem with the software. If you were paying Oracle to develop the database for you, you might have a case. But then, if that were true, you wouldn't have called tech support, would you?
This is disingenous. I have to reboot my fscking windows desktop 4 times a day at work. At work. Because the fscking PC support department just can't get it through their fscking thick skulls that they can't just push the same fscking image out to every PC. Because every batch of fscking PCs we order from the fscking vendor comes with different fscking internals even though they are the same fscking part number. Q:"Why is my PC broken again?" A:"I don't know. It worked in the lab."
On the other hand the NT servers I admin at work get rebooted when I have to patch. Period. Might be because I actually run the patch on each one instead of pushing some files and registry entries over and hoping I didn't miss anything. Then again maybe they just like me better.
At home I only reboot when some D3D game crashes my video drivers. That's hardly a Microsoft problem. It's a problem with the vendor supplied drivers.
Linux is stable. It's just irrelevant at the moment. I'd sooner go with OpenBSD on a server. As a desktop? It's a geek toy. People who just want to get things done don't use it. You know why? Because it takes longer to figure out how to get the hardware/software/user interface for a given task to work than it does to just do the work in another operating system.
This isn't directly related to the parent post but insomuch as the parent post seems to advocate Linux over Windows, allow me to relate my experience with Linux.
I've tried Linux twice. I tried it back around '97. Got the base system installed no problem. The 1024 cylinder boot limitation was annoying but not horrible. Configuring X was pretty horrible. Mostly because the kind of detailed specs X wanted about the monitor simply weren't available. PPP took me three days to get mostly working. I never did get it to consistently reconnect when the ISP did their hourly drop. X is probably the biggest reason I ended up dropping Linux. Call me a spoiled Windows user but I like being able to copy and or drag and drop without having to kill X from another box after it goes haywire. 3D gaming was pathetic. No real alternative to MS Office.
Tried it again a while ago with SUSE 7. Ran into the 1024 cylinder limitation. It was annoying on a 6GB drive. It was ridiculous years later on 20GB drive. Strike One. (Yes I know it's fixed now.) Trying to install any vaguely useful software quickly degenerated into darkest depths of conflicting libraries/ widget sets/ toolkits/ desktops/ etc. Dll Hell was never _this_ bad. Strike Two. (It isn't as bad now.) I had to recompile the kernal so it would recognize my NIC. Unfortunately something had apparently knackered gcc during strike two. (Probably fixed. I don't really know or care what the root cause was.) Strike Three.
I get paid decent money to admin computers at my job. I don't get paid to do the same at home. Time I spend trying to coerce and cojole Linux into doing the same tasks that other OS provide out of the box is less time with my family. If I'm not getting paid for it, and I wouldn't have to spend that time using a different OS, then it's a flat out waste of my time. You want Linux on the desktop? Get a version of Linux that provides the close to the same functionality of Windows or OS X out of the fscking box, and more people might take you seriously.
I really don't think the point is to boost enlistment rates. The point is raising the quality of the recruits they do get.
It takes years to turn a raw recruit into a soldier that can make good decisions with bullets flying and the plan in tatters. To do so strictly with field exercises is extremely time consuming and expensive. For small scale engagements with prep time and after action reviews, you're pushing it to get 6-10 run throughs in a day. For larger operations you might run through once or twice in a day. If a parachute drop is required, just plan on once. Period. Plus, brass isn't cheap and typically it doesn't get policed up except on a range.
A video game like America's Army allows training missions to be run at a much faster pace with little expense. It teaches cooperation, teamwork, and most importantly, tactical thinking. In other words it's a head start for the small unit leaders of tomorrow. As long as the game is realistic enough that the optimal tactics in the game carry over substantially in the real world, it's a good thing.
If ends up boosting the enlistment rate, so what? Is it a bad thing to serve your country? Is it wrong to defend your loved ones? If your objection is simply that it might make war seem cool, I can assure you that if basic training or assimilation into their unit doesn't rid them of that notion, either they aren't combat arms or they won't be for long.
It's interesting they include a PC in the tuition for the first term. $12k vs $7k for the second term. Implies a $4k-$5k price tag for said computer. Ownership doesn't transfer until after the fourth term (and a total expenditure of $29k). What a rip off.
Why not just post minimum specs each term? If a student can come up with tuition for this program, they can afford a few hundred each term to upgrade their PC if required.
Floppy Disk? By the time programs came on floppies, I was so sick of typing the programs that I was happy to pay the extra. Besides, I never did care for the noise the cassette tapes made.
Regardless of whether DRM can leverage TCPA, TCPA is an important step in the right direction for business computing. Its probably unnecessary for a consumer system. I say probably because I haven't thought of a good use other than DRM. I'm not going to make the assumption that because I haven't thought of a good use that there isn't one.
In business computing, the ability to restrict hardware and software will become a priority. For example, being able to ensure that account information was not compromised even if one of the thousands of employees with physical access to a customer service workstation installed additional software on the machine. Yes, it's against policy and they will be fired for it. Yes, we have the machines as locked down as they can be and still function. Yes, we do FBI background checks on every employee. And, yes, they can still write them down. But, if you have physical access to a computer, you can alter the operating system. If someone walks out with a file containing thousands of accounts it's a much bigger problem than a couple names and numbers scribbled on a napkin.
Aside from making it difficult to steal large amounts of sensitive information from the company, trusted hardware makes biometrics a viable form of authentication. Sure, biometrics has issues. So do static passwords. Most of the issues with biometrics can be solved through trusted hardware. The significant remaining issue, fooling the trusted hardware, can be mitigated through an additional means of authentication. Probably the smart card built into your employee ID. It isn't infallible but it's certainly more secure than passwords. Stealing an ID and forging a fingerprint or retina scan or DNA sample vs. social engineering a static password, which do you think is harder?
Frankly I don't give a damn if some software DRM application tries to leverage TCPA on my home machine. Call me old fashioned but I'm not about to pay for an ebook or an mp3 or a divx movie. What do you get? Some zeros and ones that go away if your hard drive fails? Nah..I'll pass. And if they want to stop me from ripping a CD to mp3s? Well then I can alter the hardware and/or use a different OS.
The class of games may be well defined, but this tournament is not. How long do the humans have to study payoff matrix between each move? What kind of background will the humans come from? I expect a PhD in game theory is going to play differently from a housewife.
In a simple game like this, it's possible to write a program to emulate the behavior of a particular type of person. Getting the best match on an unspecified group of people involves a large element of blind luck.
Palladium will go through. Trusted Computing will happen. It doesn't matter whether its used for DRM or not. It will happen because it provides an increased level of security and accountability for business computing.
For example, biometric authentication is basically worthless at the moment. It's all too easy to spoof, vulnerable to replay attacks, vulnerable to hardware modifications, etc. If you set up a trusted system that only accepts known hardware and software, biometrics gets a lot closer to being a reality.
As an individual, I'm not happy about trusted computing because I see all the ways it can be abused.
As a developer/sysadmin, I'm looking forward to incorporating this into our applications. No more password reset calls. No more trying to remember 15 different strong passwords without writing them down. A big fat raise because my boss won't have to remember all his passwords without writing them down.
Trusted Computing inside a business makes sense.