I totally disagree. The Cox-provided SMTP servers will transfer any outbound mail you send to them. There is never any need to use another mail server for outbound traffic. If one is using another mail server then that mail server should be requiring authentication when it receives mail from unknown IP addresses. If that is the case then that mail server should be running an MSA on port 587 in addition to or in lieu of an MTA on port 25. Therefore, blocked port 25 does not affect submission of mail to properly configured mail submission agents (MSAs).
The problem with your idea is that an unwitting user might think he has his computer secured but how can one really be certain? What does the ISP do if the user has assured it that he has no viruses/trojans but then the ISP starts getting a bunch of port 25 traffic coming from the user's machine? Do you then block the port until he clears it? Do you block all access including to the web? And there's another thing. Doing that sort of port blocking requires a slightly more advanced firewall than simply blocking port 25 outright.
If you simply block port 25 always then you never have to worry about this. And, as mentioned above and in my original post, blocking port 25 has no effect on legitimate setups whatsoever. The only thing it might do is prevent you from sending to an MTA requiring authentication in which case you are doing a mail submission not mail transfer which should be running on port 587 as a pure MSA not port 25 as a hybrid MTA/MSA.
The problem with blocking outgoing port 25 is uninformed so-called power users like yourself who claim to need it. You don't. Nobody does. Get over it and configure your shit the right way. Imagine if all of the DSL and dial-up providers blocked outgoing port 25. Can you even think about how much less spam there would be? Dial-up blacklists would become a thing of the past. Trojans would have to be smart enough to use the correct smart host. And even if they did that all of the traffic would be logged. It is an excellent idea and I can't believe that more ISPs aren't doing it.
Or you can simply block all outbound port 25 except to very specific mail servers. Cox does this. At first I was a little miffed but then I realized it makes sense. You can still send mail to anywhere you just need to go through their mail server. So if you are running your own SMTP you simply set (for example) smtp.east.cox.net as your smart host and be done with it.
This way you stop most of the mass mailing trojans because they'd have to be smart enough to use the right smart host. Then, even if they do get smart enough to do that cox still has their mail server's log so they can easily show what went out.
The only wrinkle in this is a road warrior who wants to authenticate to his company's mail server so the mail appears to be coming from there. That is simple actually. Simply run a mail submission agent (MSA) on port 587 and reconfigure the clients to use port 587. An MSA only accepts authenticated connections.
I'm not sure where I picked that up but it was definitely within my first few months of using OS X. Probably from a general OS X book. That was back in the 10.1 days and I wanted to familiarize myself with the system before porting an application to it. That's why I got the Mac in the first place. I had no idea I'd wind up using it as my main computing platform.
Anyway, the best I can dig up right now is Apple's HIG. You can learn a lot about using OS X just by reading the guidelines. Of course, there are also several books available that are written for users rather than developers.
Here's a link to the section of the HIG that covers the stuff about proxy icons and even sidebars: Apple Human Interface Guidelines: Window Appearance. You can use the path links towards the top to navigate up to the HIG root and read other parts of it. Unlike the equivalent for Windows (IBM's CUA document) the HIG is a living document. Also unlike the CUA it is designed with the Mac in mind and does not attempt to cover every type of platform in existence from mainframes to GUIs.
Replying again because I failed to address a different part of your post.
You make some valid points (what's quicksilver, btw?), and while it's possible to live without a start menu well enough, the thing that I consider to be superior about a start menu is not knowing where the apps are installed (after all, how many Windows users know where their apps are installed, but they all use the start menu), but having a hierarchical way to quickly find them. Take the KDE menu, for example. If I want to run oo.org, I click the KDE menu, point to the Office submenu, and click the component I want to run. On a Mac, AFAICT, if it's an app that's not in the dock (and then you have to know which icon it is), you have to run it from the Applications folder in Finder, or run Spotlight and type its name to find it and run it. While this is generally doable if less convenient, what happens if a user doesn't know the name of the app but knows it's (just for example), a word processor? Typing it in Spotlight has just become a non-option, and digging for it in Finder may succeed but it's harder than what I just described for the KDE menu.
This is a designed difference about the way Mac works vs. other systems. On a Mac applications are purposefully shown to the user and not hidden away in the depths of/usr/bin and/usr/share and/var and so on or C:\Program Files. The idea is that the application now becomes something the user is able to grasp. Internally, applications are actually folders with a particular structure. If you are curious you can either view them with the command-line or you can use the Show Package Contents menu option. This is available either by right-clicking (see other post for what this means) or by using the great contextual menu in the sky (i.e. the menu bar which is context-dependent on a Mac) and choosing Show Package Contents from the File menu in Finder with the application icon selected. Oops.. except for some stupid reason Apple didn't put it there.. arguably it should be. Maybe I'll file a bug.
One more thing, regarding virtual desktops. Are you aware that Spaces has been announced for Leopard which is a really good implementation of them?
BTW, would you happen to know how I can get my applications folder visible again? It vanished after I dragged it to the dock as the AC below suggested. Also, he mentions right-clicking things in the dock. How does one do that with a one-button mouse (trackpad)?
I'm guessing you dragged the Applications folder from the left side of a finder window. Unfortunately, a drag from there to anywhere outside of it simply causes it to poof away. That left pane on a finder window is very much like the Dock. It doesn't represent a folder or file on disk but instead is a reference for it. Yes, it's an oddity I'm not too happy about but on the other hand if I could think of a better idea I would.
At any rate, to get your applications folder back simply switch to Finder then select Go->Applications (which is Cmd+Shift+A). That will take you to it. Now, you can do one of two things. The cool way (for those in the know) is to click and hold on the proxy icon which is the icon in the titlebar of the window. After about 2 seconds it will shade darker. You are now dragging the folder itself. It's a nifty mac feature I really wish Windows and Linux had. The alternative is to go up a level (Cmd+Up or Go->Enclosing folder) to the parent folder (which in this case is root of the drive) and drag the Applications folder from there. This of course matches the Windows/Linux behavior. Anyway, simply drag it into the left-pane and you'll recreate the reference to it in the left-pane again. Then drag it again onto the right side of the dock and you'll have created a reference to it in the Dock as well.
To bring up a menu showing the contents of a folder in the dock simply bring up the contextual menu for it. You can do this either by using the right mouse button (if you have one) or by holding control and using the left button or in the case of the dock by simply holding down the left button without moving for about 2 seconds. A final alternative is to configure your trackpad to respond to a tap with a single finger as a left click and a tap with two fingers as a right click. You can find this in the Mouse preference pain in the System Preferences application.
Also note that it is my understanding from the documentation that comes with new macs that for the first 90 days of Mac ownership you can actually call apple and get answers to these types of questions. If you buy AppleCare you extend that to 3 years. Of course if you got your Mac from work then the IT department there may or may not know what the hell they are doing with Macs.
One last thing just to add a quick "cool factor" to OS X. Remember that proxy icon in the title bar? Hold down Cmd (i.e. the key with the apple logo and the swedish campground symbol on it known as the "Command" key) and click that proxy icon. Notice that the menu that pops up shows you the file and all of the enclosing folders. This also works for documents you have open in applications like Word or Pages and what not. Good luck with your new Mac!
Great, another OO.org fanboy talking about how he swears he's going to use OpenOffice.org more and so are other people because Office 2007's user interface is just too drastically different. It just seems obvious that people want the same interface they are used to, why change it?
I haven't used Office 2007 myself but what I am seeing in the ribbon interface is merely docked palettes. Look at any decent Mac application, particularly anything using Cocoa or coming from NeXT (see Omni Group programs) and it will typically use palettes as the primary interface for interacting with object properties. See OmniGraffle or OmniOutliner or OmniPlan. Or see Apple's Pages and Keynote applications.
All Microsoft has done here is reduce the menubar to the file menu and replace the toolbar with what essentially amount to palettes and drawers as found in NeXT/Cocoa applications. NeXT did this 10+ years ago. However, Microsoft has stuck with the old menubar/toolbar interface. Microsoft's menubar is just a bastardization of Apple's which removed it from the top of the screen as a global contextual menu and put it inside the window as a hard to point at set of choices. Microsoft's toolbar is I believe a Microsoft UI invention that started off with the good intention of making a small button for each of the most common functions of an application.
Microsoft's ribbon interface has finally shown that contextual interfaces are the best UI paradigm we know of today. For what it's worth, Office 2004 and v.X for Mac both have palettes although sadly maintain the modal dialogs as well. I really hope their MacBU team does a good job with the next version of Office for Mac.
Could they have done better? Well maybe, I haven't used it yet so I don't know. I will say the default blue-ish color scheme is garish to say the least. However, it does look like Microsoft is finally starting to realize that adding cute animated characters to a poorly designed interface doesn't help. No one really wants to learn how to use a program, they want a program to work as they would expect it to.
What interests me more is how they were able to pull this off from the programming side of things. The Win32 event and windowing model does not lend itself well to these types of interfaces. It would take some smart developers to come up with the necessary framework to implement this type of interface. In contrast, Cocoa's target/action system makes writing these types of interfaces a breeze. The OpenOffice.org team has one hell of a task ahead of them now.
Still, OO.org hasn't been all bad. I think its biggest contribution to office software is hilighting how completely crappy Microsoft's Office interface has been and showing Microsoft that if they want to compete they are going to have to do better. My guess is that it's going to be several years before OO.org catches up unless they move to some sort of toolkit that does this stuff for them. If I were them, I'd ditch the crap they have now and move to GNUstep and put some development time on it. Of course, it would be blasphemy to use Objective-C at Sun since Java is supposedly the evolution of Objective-C. Having used both I can say it seems to me that Java is a step backwards not forwards. The reality here is that OO.org is basically dead because Microsoft just killed it the old fashioned way: by making a better product.
FYI, no one is going to consume less. Not going to happen. Ever.
However, if you want to take a first step I have a modest proposal: simply kill yourself. You will use much less energy that way. That is, after all, the logical extreme of reducing energy consumption.
Several companies do intra-company billing. Not all, but those that don't probably should. As shown by this discussion thread it's killing your business if you don't. Simply replace "that's not a supported configuration" with "this is a problem specific to your configuration. Do you have a charge number?" and things will move along much more smoothly.
Of course, some idiot managers will shit when they see how valuable your time is and swear up and down that they can get it cheaper elsewhere. And they can go right ahead and do that. Assuming you're running your IT department as a tight ship they'll come crawling back to you real soon. After, of course, blowing much of their budget on some consultants.
Now, granted not all of IT is direct user help. Sometimes IT does for itself to make its job easier and sometimes IT works in general for the whole company to improve infrastructure. However, when IT is working to help a specific user or a specific department then it needs to be accounted for that way.
This is normally referred to as a "Service Center". And how do I know this stuff? Because I'm a developer and general IT worker (it's a small company) who does work on internal accounting software. I'm fortunate enough though to be in a place where our accountant organizes things as if we were a large company who wanted detail on every cost. This is particularly important because we are a mid-sized government contractor. When I work to improve infrastructure I work on one charge number. When I work helping some id10t user with their problem I work on their money; I don't mind, I'll take all the time they need to understand it-- within reason of course because there is still opportunity cost involved. I certainly NEVER refer to them as an idiot to their face or around the office (gossip spreads!!!) and I attempt to understand what they are doing wrong or what our systems could do better. When I develop software for accounting for costs on a specific contract I charge to that contract's overhead. If I write general accounting or business management software I have yet another charge number to apply it to G&A.
It's not hard at all for me to do this and it gives everyone the raw data they need to come up with an accurate idea of where money is going. It keeps our costs low and keeps our business efficient. I have to say if your business isn't doing it then you need to get a better accountant and maybe get or write better accounting software. If you're not doing real timesheets indicating how much you worked each job on each day then start.
Presumably, if you already have a copy center that functions this way, you probably already have at least some of the accounting infrastructure you need.
I will give you that when you are working on your own overhead that it quickly becomes clear that supporting all sorts of different configurations takes a lot of resources. However, if you are able to charge to the user's job then it will quickly become apparent to the users (or at the very least their managers) that your time is valuable.
Unfortunately, many companies simply do not have their accounting and timekeeping systems set up to handle this situation. In effect they are letting their accounting software dictate how the business is run. It's a shame really but hey what are you going to do? You're IT. That's not your problem, right?
How is the imperial system is "handicapping" us? If anything, most educated Americans are able to work in units of either system. And as you pointed out, it's not as if we don't decimalize where appropriate. In addition to your machine shop example a more common example would be gun calibers. I'm sure everyone has heard of a.22 ("twenty-two") or a.357 Magnum ("three fifty-seven Magnum") or a Colt.45 ("Colt fourty-five") or a.50 caliber Desert Eagle ("Fifty caliber Desert Eagle"). All of those are expressed in inches. For reference, a 9 mm would be about a.354 caliber gun (simply convert 9 mm to inches).
All of those are in inches but nobody says a "point two two" or "point three five seven" or any weird crap like that. Which is, I think, yet another example of how the imperial system caters more to speech and common units. No one in their right mind says 2 feet 3 and 3/4 inches. If you really _are_ measuring down to 1/4 inch precision then it is 27 and 3/4 inches. Now say you are doing this in metric. Is it 70.5 centimeters? In that case, probably yes. But what if you're measuring 74 and 1/2 inches? Note that that is 6 feet 2 and 1/2 inches but no one would ever say that. Is that now 189.2 centimeters? Or is it 1.892 meters? From an american perspective the centimeter measurement makes more sense to me since that is the unit I'd most likely be measuring. But god knows what the crazy rest of the world does here.
I also want to reiterate your point about fractions. Halves and thirds and fourths and so on are extremely common divisions in everyday life. It's nice to know that a gallon has 4 quarts and a quart has 2 pints and that there are 2 cups in each of those pints. Nice simple division by powers of 2. Should make a lot of sense to a bunch of computer geeks wouldn't you think!? Interestingly enough, all of those can be represented without problem in a binarialized (is that a word?) system. Computers do in fact store floating point as ?/2, ?/4, ?/8, ?/16 and so on with the exception being most good desktop financial calculators which are explicitly designed in hardware to store in tenths and hundredths and so on to avoid rounding errors on financials which are now all decimalized. Oddly enough, the old pieces of 8 system made a hell of a lot more sense from the perspective of trying to store it as a floating point number in a computer.
My guess is that metric is going to take off in the U.S. about as well as Esperanto has. Languages and measurement systems are born out of the necessity for peoples to communicate, not through government mandate under some grand socialistic scheme.
Re:You have no idea what you're talking about
on
SQL Hacks
·
· Score: 1
Have you had a look at Apple's WebObjects?
It is a completely different beast from something like Hibernate or even Rails. Like other frameworks it maps database tables to entities (object classes) and database rows to instances of those classes and database columns to accessors in the class. And like other frameworks it is able to model relationships from one entity to another. And of course it does pull in data and cache it in application memory. Of course, any database access is bound to do this at some point.
However, unlike some persistence frameworks it does not attempt to enforce a hierarchical view of the data. In fact, it truly is a relational system. The big difference with WebObjects is the concept of the editing context (EOEditingContext class). The editing context is responsible for tracking all changes to the object graph. It is capable of working with hierarchies of course but it is also capable of a whole lot more.
Essentially, as the user modifies objects and modifies relationships between objects the editing context keeps track of what is going on. Ultimately, when saveChanges is called on the editing context (NOTE: NOT on some object) all of the changes are batched into the database within a transaction block. That may cause one object to be modified, another to be inserted, some to be deleted, or any combination of those.
In general, inserts are not going to conflict because one typically uses a database sequence to generate integer primary keys or is using compound PKs with enough unique user-entered information that they would not conflict. Should they conflict, EO is able to trap this. With updates, EO defaults to an opportunistic locking scheme whereby update statements are qualified with all attributes of the object (not just the primary key) such that if the record has been changed in the database the program can trap it. There is a facility for defining what merge behavior is appropriate for the particular entity by implementing it in a method that will be called in the event this occurs.
Also, unlike some other frameworks it is able to model EVERYTHING you can put in a SQL database. Or a non-SQL database for that matter! I have quite successfully used it to access a (copy of a) number of COBOL files as if each COBOL file was a table in a relational database. It helps of course that the COBOL data store was sort of relational to begin with of course. Where COBOL arrays were used I have successfully defined views in PostgreSQL to unwrap these into a set of related tables which I then model in EOModeler.
I'm not saying that the Enterprise Objects part of WebObjects is perfect, but it's as close as I've seen. The reason it is so good is that it truly follows the relational model and if you really think about it it sort of is an RDMS in and of itself. Think about the act of opening an editing context then either calling saveChanges() or revert(). Does this sound like "BEGIN" followed by "ABORT" or "COMMIT" to anyone? Of course, it goes beyond that since it's doing all of this in its own memory. Gets even better when you nest editing contexts. Suppose you have a deep hierarchy of objects. Yes, I realize we're dealing with a relational system, but sometimes you really do want hierarchy. In an app using EO the user is able to drill down from the root class and change all sorts of things about it and its children and its children's children and so on. Only when the user saves back up to the parent does it commit all of this in one big block. Primary keys and foreign keys are taken care of based on the rules you specify.
It really is a pretty damn amazing system. If you admire the relational data model I assure you you will like EO. Even better, model some DB you have and point D2W at it. It is absolutely amazing. Unfortunately the UI was designed by engineers in the early to mid 90s so it does tend to look like shit. But it WORKS. And looks are easy to change. Probably EOs biggest downfall is that it does so accuratel
My first experience with a computer was one that I programmed in BASIC at the age of 5. I'm sure there's plenty of other Slashdotters with a similar experience. It's not unreasonable to think that some children in a third-world country might have the talent for programming.
The only problem I see is that without access to decent tutorials and documentation it is hard to learn programming. Perhaps what's on the internet will be enough but perhaps not. In my case I had the Hands-on BASIC for the IBM PCjr (ISBN: 0070491690) book as my starting point. Wish I still had a copy of it. I remember it being really easy to read and follow and I was a child at the time.
I noticed that a comparison to Objective-C is quite conspicuously absent from the list of languages compared to D. Why is it missing? Granted D seems to be a much greater change to C than Objective-C is but I can't help but thinking that one of the main attractions to D seems to be its heap-based garbage-collected object system. You can already get the object runtime with Objective-C. If you use GNU you can even have Boehm GC (which is apparently the GC that D uses). If you use Apple you will have to wait for Leopard to get GC. Another new Objective-C feature is the ability to use full C++ objects as instance variables in your Objective-C classes and do the right thing with initializing (calling the default no-argument constructor upon alloc).
On top of that, Objective-C actually includes tons of reflection information. Although Objective-C has protocols which are roughly equivalent to Java/C# interfaces they are almost completely unnecessary. In Objective-C one can query at runtime whether a method is implemented or not and if so call it. So whereas in Java you'd do this:
The difference being that in the Java case you have to declare MyInterface as containing the one doSomething() method and inform java that your object extends MyInterface whereas in Objective-C you merely need to provide a doSomething method on your object.
Basically that means that in Objective-C every single method effectively becomes an interface. You would not believe how useful this is once you realize it. Note that at runtime there is ZERO difference. In both the Java and Objective-C cases the object is being checked to see if it implements something. Same with C++ if you use dynamic_cast<>()
Granted every language has its niche and I'm sure D will find its. Objective-C's niche is definitely GUI programming. The ample reflection information allows for easy implementations of archiving (serialization) and most importantly key-value coding and the related action methods pattern. It's a pretty damn cool thing when your RAD tool simply outputs archived objects that refer to methods to be called upon certain actions simply by name.
You can give somebody backdated options only if you debit your company account by the current cost of the options.
No, you credit a liability account. Regardless of the account type credits are always negative and debits are always positive. On a liability account a credit, being negative, increases liability. Unfortunately, under the traditional accounting system negatives are never used and thus it is recorded as positive credit because it is an increase. But it is an increase on an account which naturally has a credit balance and thus it is effectively negative.
Further confusion arises because when your bank prints your statement and talks about your account they refer to their liability to you. Thus, when you deposit a check they credit "your account" and when you withdraw money they debit "your account". In reality, it is not in fact your account but is rather their account of what they owe you if you decided to withdraw all of your money immediately.
Obviously they don't ever let you have a debit (note: positive) balance because then you'd owe them money. They always want to owe you money because that is the business they are in.
This is perhaps more readily apparent if you come up with a more similar transaction. Say you earn interest on money in your savings or checking account. You would of course expect them to credit "your account" for the amount of that interest. This is a good example vs. deposits/withdrawls because it's not a transaction you initiated. Now, say a company gives an employee some stock options. What they are doing is giving the employee money. So, wouldn't you expect them to credit their ledger?
Another way to think of it is that if you have a debit balance on your ledger (say you are a banker) then that means someone is in debt to you. Or you could say that if the banker has a debit balance on his ledger then you are in debt to the banker. The etymology of debit reflects this.
Now, let's get back to the stock options. Put on your corporate accountant hat. You (the company) has granted options to an employee. This means the employee has the option to buy up to x shares of stock at the price it was when you granted the options (or possibly when you declare the options have been granted). When at some unknown time in the future this transaction occurs the employee is going to give you cash. You know this amount in advance because you have fixed the price the employee will have to pay for the stock. What you don't know is what you'll have to pay (an expense) to buy this stock for the employee. What price will shareholders charge and thus what will your cost for that stock be? When the transaction ultimately occurs it is two transactions. One will take money from the employee and debit (increase the value of) some cash (equivalent) asset account. The other will cost you money and credit (decrease the value of) some cash (equivalent) asset account. Since the employee is unlikely to exercise options that cost him money you can be sure that ultimately it will cost you money resulting in a credit to your net worth. But how much and when? Who knows!
It is at this point where I have to leave it because I am not an accountant but rather a programmer. All I can tell you is that there are various accepted models for determining the price you will probably have to pay for the shares in the future. That is how you can determine your effective liability. It is very much a "guessing game" since the best you can do is use some probabilistic method.
This is all assuming of course that you do want to record the liability. There is also some chance that the options will expire and never be exercised! It's not so much that I'm not an accountant but more that I'm not a fortune teller.
Ultimately, a stock option grant results in an increase of projected liability. But in recent years some people want to see this accounted for not as an increase in projected liability but
RIM didn't design the cell phone network, dorkwad. RIM the other devices work over the approved and paid for data channels. If you are paying $80/mo for "unlimited data" you better damn well get what the system can handle. And FYI, CDMA and GSM systems use a time division multiplex method so it's not like a cable modem link shared with the neighbourhood; you can't choke the whole system by being a data-hog, at most you can consume your alloted time-slice.
As other replies have mentioned, CDMA stands for Code Division Multiple Access. Each packet is tagged with a small code similar to the way ethernet packets are tagged with a MAC address. GSM is a TDMA system (Time Division...) which relies on phones transmitting in an allotted time slot (milliseconds). This could be likened to a token ring network in some ways
That is bad enough but then you go on to state that it's not like a cable modem link. Huh? It is exactly like a cable modem link. In fact, just about any network or bus, wired or wireless, uses some sort of scheme to allow for multiple transmitters. In DOCSIS you have both a MAC (code) as well as an allotted time slot. DOCSIS takes it a bit further and does ranging. Basically, it pings the node at the head end until it gets an accurate picture of how long an electrical signal takes to reach the node. This allows the cable modem to adjust its transmissions such that when it transmits in a given timeslot it won't collide with another modem's time slot due to signal delay differences. The speed of electrons is after all only so fast.
I could go on and on here by pointing out that for example a T1 line is actually 24 time divided channels or that a PCI bus works on an acquisition basis. Next time before calling the parent poster a dorkwad (what are you 11?) you might actually want to know what you are talking about.
Stop voting the party line and vote for whom ever you think will do the best job, be it Democrat, Republican, Libertarian or an Independent candidate.
This is very good advice and I recommend everyone heed it when voting tomorrow. In general I tend to vote republican because I like their positions on issues. I am in favor of people getting paid what they will take, not a predetermined national minimum. I am in favor of health insurance not nationalized health care. I am in favor of giving parents the economic freedom they need to choose where they send their children to school, not forcing them to use a failing public school. I am in favor of allowing the executive branch to intercept communication with one endpoint outside the U.S. without a warrant. I am in favor of parents being notified if their daughter is going to have an abortion just as they would be notified with any other medical procedure. I believe a woman who shoots herself in the stomach two days before delivery should be charged with murder. I believe abortion mere days or even weeks before expected birth date is clearly murder.
On the other hand, I'm in favor of allowing gay couples to make contracts providing marriage-like benefits. I am in favor of legalizing abortion legislatively, not through some assinine court decision involving a legal house of cards. I would hope though that we could have some working sex education that explains to kids both the proper usage of condoms as well as pushes the idea that abstaining from sex is a wise decision for someone not prepared to be a father. Of course, that leads into a whole new political quagmire.
You see, I simply believe that a man should stand on his own but that a man's community should offer him help to get back on his feet when necessary. I do not believe in giving welfare checks to people for having babies. I do not believe in shoving the kids into public schools then lowering the testing standards so everyone can "pass".
I could go on and on but I'll stop here. What does that make me politically? I guess more a republican than a democrat. Although not entirely because there are a number of good democrats with views very similar to mine since these are I believe rather moderate views.
I am in Virginia's 2nd district. That means I get to vote Drake (R,I) vs. Kellam (D) and Allen (R,I) vs. Webb (D). Unfortunately, most of what I've seen from both sides is a negative campaign. I have seen and heard _some_ issues talked about by the democrats and some talked about by the republicans. And when I ignore all of the negative ads and focus on the issues ones I find I like the positions of the republican candidates more.
The thing is though, go to say Phil Kellam's website. Can you find where he stands on any issues? Hmm. Neither could I. His website has been this same thing primary attacking Drake and highlighting his great work banishing city stickers in Va. Beach. I will say his website looks "slick". Sort of like he does. Clinonesque I guess. The only place I can find anything is the "Media Center" section. Even there I find crap about Foley and the GOP did this and Drake did this and so on and so forth.
Drake's website isn't much better. The best section is again the "News" section which is showing news stories from local papers. Drake's site still has an awful lot about her challenger but has at least some about what drake stands for. Not as much as I'd like though
Do I like the way things are going in Washington? Not entirely. Am I going to go out and vote Democrat to send those damn Republicans a message? FUCK NO. What message would I be sending to washington? Run a better smear campaign and don't ever talk about the issues at all and you can win? Great, next election cycle we can expect more of this shit because it obviously worked!
How things will come out tomorrow is anybody's gues
I think the reason you stopped using multiple monitors on your Mac is that the menu bar only shows up on one of them.
Dom, I think you hit it spot on with respect to multiple monitors. I think you have it wrong about larger monitors though. Even on my 23" display the menubar is close enough (only need to bang the mouse up there with a quick flick of the wrist) that it's still highly usable. Not to mention that being the geek I am I tend to remember shortcuts for things I do often. Also, usage of the menubar can be reduced when drag and drop is available.
Similarly, people don't work with maximized windows on Macs because there is no maximize feature. There's a "zoom" button (the green blob, I think), which either makes the window bigger or smaller, but it doesn't maximize. In my experience people like to eliminate distractions.
I think the reason for not having maximized windows is that it is a multitasking computer and that going in to single-task mode is highly ludicrous for your typical workflows. With that said, the iApps tend to be designed around the one maximized window model. It seems to work for those because they are like their own self-contained bottles (i.e. this is the microsoft model). I don't think that people really want to eliminate distractions. I think that some people are just neat freaks. Of course, using "Hide Others" from the application's menu (the one in bold) will handily eliminate all distractions. Assuming, of course, you don't have a billion icons on your desktop.
With spaces in leopard eliminating distractions will be even easier because one can simply configure a space with only the things they're working on. This, I think, is Apple's answer to the good things about maximized windows. That is, that they eliminate background distractions.
Part of the Mac model is that the menubar switches with the app you're using...
You just hit on the one thing that I hate most about OS X (and the rest of the Apple OS line, all the way back to the original Macs). Apple OS's are the only ones that have ever (AFAIK) done this. My Gnome menu does not change when I open GIMP. The Start menu in XP doesn't disappear when you bring up WMP. Why in Christ's holy name would you want it to?
The best way I can explain it is that the menubar on OS X is not much at all like the menubar on other systems but much more like the contextual menu (i.e. right-click menu) that other systems have. It follows what you are working on. As you change apps all of the top-level labels change. As you move around between things within an app items in the menus are disabled or enabled as needed. Of course, OS X also has contextual menus but because Apple has a full menubar with all of the commands in it the contextual menus tend to only have a few well used ones.
If you stop thinking about the menubar as controlling the window and start thinking about it as controlling the specific thing you're working on then it makes a lot more sense. It's funny you should bring up The GIMP since it actually uses the right-click on image gets a menu model. And it's not your typical contextual menu but rather a replacement for a menu bar. That model is almost exactly the Mac model except the menu moves to the top of the screen horizontally and all the top level labels are visible without having to first right-click. Apply that to the entire OS and maybe you can see why it's not such a bad idea after all.
Of course, it does have its problems. If you twitch-click on your desktop or inadvertently miss the menubar by a few pixels you will change the menu to the Finder's menu. To me the simple solution here is to get rid of the desktop entirely. NeXT for instance did this. NeXT didn't have a common menubar but clicking (left mouse button actually) on the desktop brought up the menu which is essentially a GIMP style menu (or is it the other way around?) for the currently active window.
The desktop metaphor sort of made sense in the early Mac days when it was showing basically what was on the currently inserted floppy disk. You double clicked something and it came back to life as a document. Now that we have much more storage space the whole idea of the desktop is broken. I am all for the NeXT model of making the desktop into a workspace where only windows exist and making the file browser just a normal app.
Windows DOES have a Cut command in Explorer, something that still boggles my mind about the Mac (how can Finder not have a Cut?)
I believe it's actually a design decision on the part of Apple. The traditional way to move or copy files on a Mac has always been to use the mouse to drag them. This isn't hard at all when you have a decent sized screen and you can simply stagger the source and destination windows then drag from one to the other.
It is interesting though because dragging files is really something someone needs to be shown. My experience has been that people don't just pick it up without at least some minor prompting. Once you show someone on a Mac they seem to understand it quickly. However, I've had a hell of a time showing how to do it on Windows PCs. It just seems that people can't get their mind out of the one maximized window mindset and it's rather hard to drag from one maximized thing to another. Of course, you can drag through the task bar but that's another learned behavior, one that doesn't make that much sense compared with a normal drag.
This, I think, is one of the major shortcomings of Windows. Microsoft has basically crippled the UI to the point where it's nearly impossible to run more than a few apps with more than a few windows open. Unfortunately, it seems that Vista doesn't really fix this shortcoming. They have a cool looking alt-tab replacement but it's just that, cool looking.
It would be very hard for Microsoft to move to the Mac model here. Part of the Mac model is that the menubar switches with the app you're using and that all the toolbars and pallets disappear when the app is not active and switch when you switch which document you're working on within an app. Contrast this with the Microsoft style of putting giant sidebars on all four sides of the document area within the window. It makes the windows too big to be sized anything other than maximized on many screens.
Of course, some people have a preference for the Windows way. They say it "looks cleaner" because they only see what they're working on. Maybe some people really get distracted by having portions of other windows behind their active one still visible. Funny enough, that aspect of OS X never bothered me. I found it relatively easy to get used to the idea that windows generally exist on the screen and don't try to own the entire screen. To me it seems similar to the way one stack of paper sometimes obscures another on my real desk. I never stack everything neatly in piles and grid them out like tiles. I've got one pile of papers that's half covering another so I can see at least part of what's under it to know it's there. This way I can put a lot more crap on my desk and still know where it is. Now I know I'm not the only person whose desk looks this way
Still, can I really blame Microsoft for these things though? Not really. They made these decisions years ago trying to get people to move from DOS to Windows and then later from Windows to newer versions of Windows. The latest trend I'm seeing is for some people to get dual monitors on Windows. This way they can have two apps maximized, one maximized on each screen. I ran dual monitors on OS X for a while but lack of real maximization (and no desire to have it either) means you wind up with a good sized worksurface with a huge line in the middle of it. I've since decided that Apple is defiitely on the right track with the bigger displays. Particularly if you have the 23" you can begin to see how it completely changes how you want to interact with the computer. You're not going to maximize things; even at the smaller 20" size a window would be ridiculously big. What I find myself doing is just staggering more and more windows all over the place. It looks just as messy as my real desk. This, I think, is exactly the point. Apple has taken the desktop metaphor one step further with these huge displays.
And what has Microsoft offered us? More of the same. Compu
It seems fundamentally impossible to make a language like Python or Ruby fast. By their very nature everything has to be done at the last minute, based on string comparisons, and you can't do global optimizations really because at any moment the program might change itself and invalidate them. Consider the way every object can implement a fallback method that is called if somebody invokes "foo.bar" and bar does not exist in foo. It implies that every single method invocation must be identified by string not a number, and matched by string comparison.
It doesn't have to be this way. Take Objective-C for an example. All messages are identified by a selector which is actually just a plain old C string. Take for example [NSView -frame] which returns an NSRect giving the frame (window) size and location. FYI the syntax in objective-C to call it is NSRect frameRect = [someView frame]; If it were more C/Java like syntax it'd be NSRect frameRect = someView->frame();.
Anyway, getting back to the point: The way the compiler compiles this is a call to objc_msgSend(someView, ___selector_for_frame___) where ___selector_for_frame__ is simply a C string. Recall that C strings are simply pointers to an array of char. What the compiler does is unique all references within a single compilation unit (a.o file) such that all uses use the same pointer. Furthermore, the link editor (ld) sets things up such that they all point to the same thing within the executable or library or framework you are linking. Then, the dynamic linker ensures this when it loads in dynamic code. This way, the runtime need only do a comparison of the pointer itself (where in memory it is pointing) rather than checking what it's actually pointing to.
Basically, that is to say that the test someSelector == someOtherSelector actually works. It is not necessary for the runtime to do strcmp(someSelector, someOtherSelector) == 0.
I assume similar tricks are used in other dynamic languages. I only mention Objective-C because it seems to be the one with a bunch of publicly available documentation on its inner workings.
That's what the tabs are for on the undersides. To me it just came naturally to put the tip of my thumb on the left tab and the tip of my ring finger on the right while holding the button down with both my index and middle finger.
The mouse is light enough it hardly takes any effort to do this at all.
It's the same size as Apple's Pro mouse, which they've been shipping since about 2000 or so (it's gone through a couple of minor revisions, but the same basic size/shape). It works for most people.
I have larger hands than most people (though not compared to my size since I am 6'6"). The Apple mouse has been great for me. The thing about the Apple mouse is that you don't grip it like other mice. I don't (can't actually) move the mouse around on the pad with my hand squarely on top of it. What I do is rest the palm of my hand (except the heel) on the top of the mouse. I let my thumb and pinky rest on the mouse pad. My 3 middle fingers curl over the top of the mouse. The heel of my hand naturally curls over the back.
Basically, it is very much like grabbing a ball. The other thing is that you will notice Apple's accelleration curves are fairly slow. Because of this, you tend to need to pick up the mouse from time to time to do a drag across the screen. The way the Apple mouse is shaped it's very easy to do this since you just sort of squeeze it. Unfortunately, some idiot at Apple decided to make that an action so every so often I inadvertently activate Exposé.
Other than that, the Mighty Mouse and the Pro Mouse before it have been great.
Slashdot Mirror
I totally disagree. The Cox-provided SMTP servers will transfer any outbound mail you send to them. There is never any need to use another mail server for outbound traffic. If one is using another mail server then that mail server should be requiring authentication when it receives mail from unknown IP addresses. If that is the case then that mail server should be running an MSA on port 587 in addition to or in lieu of an MTA on port 25. Therefore, blocked port 25 does not affect submission of mail to properly configured mail submission agents (MSAs).
The problem with your idea is that an unwitting user might think he has his computer secured but how can one really be certain? What does the ISP do if the user has assured it that he has no viruses/trojans but then the ISP starts getting a bunch of port 25 traffic coming from the user's machine? Do you then block the port until he clears it? Do you block all access including to the web? And there's another thing. Doing that sort of port blocking requires a slightly more advanced firewall than simply blocking port 25 outright.
If you simply block port 25 always then you never have to worry about this. And, as mentioned above and in my original post, blocking port 25 has no effect on legitimate setups whatsoever. The only thing it might do is prevent you from sending to an MTA requiring authentication in which case you are doing a mail submission not mail transfer which should be running on port 587 as a pure MSA not port 25 as a hybrid MTA/MSA.
The problem with blocking outgoing port 25 is uninformed so-called power users like yourself who claim to need it. You don't. Nobody does. Get over it and configure your shit the right way. Imagine if all of the DSL and dial-up providers blocked outgoing port 25. Can you even think about how much less spam there would be? Dial-up blacklists would become a thing of the past. Trojans would have to be smart enough to use the correct smart host. And even if they did that all of the traffic would be logged. It is an excellent idea and I can't believe that more ISPs aren't doing it.
Or you can simply block all outbound port 25 except to very specific mail servers. Cox does this. At first I was a little miffed but then I realized it makes sense. You can still send mail to anywhere you just need to go through their mail server. So if you are running your own SMTP you simply set (for example) smtp.east.cox.net as your smart host and be done with it.
This way you stop most of the mass mailing trojans because they'd have to be smart enough to use the right smart host. Then, even if they do get smart enough to do that cox still has their mail server's log so they can easily show what went out.
The only wrinkle in this is a road warrior who wants to authenticate to his company's mail server so the mail appears to be coming from there. That is simple actually. Simply run a mail submission agent (MSA) on port 587 and reconfigure the clients to use port 587. An MSA only accepts authenticated connections.
I'm not sure where I picked that up but it was definitely within my first few months of using OS X. Probably from a general OS X book. That was back in the 10.1 days and I wanted to familiarize myself with the system before porting an application to it. That's why I got the Mac in the first place. I had no idea I'd wind up using it as my main computing platform.
Anyway, the best I can dig up right now is Apple's HIG. You can learn a lot about using OS X just by reading the guidelines. Of course, there are also several books available that are written for users rather than developers.
Here's a link to the section of the HIG that covers the stuff about proxy icons and even sidebars: Apple Human Interface Guidelines: Window Appearance. You can use the path links towards the top to navigate up to the HIG root and read other parts of it. Unlike the equivalent for Windows (IBM's CUA document) the HIG is a living document. Also unlike the CUA it is designed with the Mac in mind and does not attempt to cover every type of platform in existence from mainframes to GUIs.
Replying again because I failed to address a different part of your post.
This is a designed difference about the way Mac works vs. other systems. On a Mac applications are purposefully shown to the user and not hidden away in the depths of /usr/bin and /usr/share and /var and so on or C:\Program Files. The idea is that the application now becomes something the user is able to grasp. Internally, applications are actually folders with a particular structure. If you are curious you can either view them with the command-line or you can use the Show Package Contents menu option. This is available either by right-clicking (see other post for what this means) or by using the great contextual menu in the sky (i.e. the menu bar which is context-dependent on a Mac) and choosing Show Package Contents from the File menu in Finder with the application icon selected. Oops.. except for some stupid reason Apple didn't put it there.. arguably it should be. Maybe I'll file a bug.
One more thing, regarding virtual desktops. Are you aware that Spaces has been announced for Leopard which is a really good implementation of them?
I'm guessing you dragged the Applications folder from the left side of a finder window. Unfortunately, a drag from there to anywhere outside of it simply causes it to poof away. That left pane on a finder window is very much like the Dock. It doesn't represent a folder or file on disk but instead is a reference for it. Yes, it's an oddity I'm not too happy about but on the other hand if I could think of a better idea I would.
At any rate, to get your applications folder back simply switch to Finder then select Go->Applications (which is Cmd+Shift+A). That will take you to it. Now, you can do one of two things. The cool way (for those in the know) is to click and hold on the proxy icon which is the icon in the titlebar of the window. After about 2 seconds it will shade darker. You are now dragging the folder itself. It's a nifty mac feature I really wish Windows and Linux had. The alternative is to go up a level (Cmd+Up or Go->Enclosing folder) to the parent folder (which in this case is root of the drive) and drag the Applications folder from there. This of course matches the Windows/Linux behavior. Anyway, simply drag it into the left-pane and you'll recreate the reference to it in the left-pane again. Then drag it again onto the right side of the dock and you'll have created a reference to it in the Dock as well.
To bring up a menu showing the contents of a folder in the dock simply bring up the contextual menu for it. You can do this either by using the right mouse button (if you have one) or by holding control and using the left button or in the case of the dock by simply holding down the left button without moving for about 2 seconds. A final alternative is to configure your trackpad to respond to a tap with a single finger as a left click and a tap with two fingers as a right click. You can find this in the Mouse preference pain in the System Preferences application.
Also note that it is my understanding from the documentation that comes with new macs that for the first 90 days of Mac ownership you can actually call apple and get answers to these types of questions. If you buy AppleCare you extend that to 3 years. Of course if you got your Mac from work then the IT department there may or may not know what the hell they are doing with Macs.
One last thing just to add a quick "cool factor" to OS X. Remember that proxy icon in the title bar? Hold down Cmd (i.e. the key with the apple logo and the swedish campground symbol on it known as the "Command" key) and click that proxy icon. Notice that the menu that pops up shows you the file and all of the enclosing folders. This also works for documents you have open in applications like Word or Pages and what not. Good luck with your new Mac!
Great, another OO.org fanboy talking about how he swears he's going to use OpenOffice.org more and so are other people because Office 2007's user interface is just too drastically different. It just seems obvious that people want the same interface they are used to, why change it?
I haven't used Office 2007 myself but what I am seeing in the ribbon interface is merely docked palettes. Look at any decent Mac application, particularly anything using Cocoa or coming from NeXT (see Omni Group programs) and it will typically use palettes as the primary interface for interacting with object properties. See OmniGraffle or OmniOutliner or OmniPlan. Or see Apple's Pages and Keynote applications.
All Microsoft has done here is reduce the menubar to the file menu and replace the toolbar with what essentially amount to palettes and drawers as found in NeXT/Cocoa applications. NeXT did this 10+ years ago. However, Microsoft has stuck with the old menubar/toolbar interface. Microsoft's menubar is just a bastardization of Apple's which removed it from the top of the screen as a global contextual menu and put it inside the window as a hard to point at set of choices. Microsoft's toolbar is I believe a Microsoft UI invention that started off with the good intention of making a small button for each of the most common functions of an application.
Microsoft's ribbon interface has finally shown that contextual interfaces are the best UI paradigm we know of today. For what it's worth, Office 2004 and v.X for Mac both have palettes although sadly maintain the modal dialogs as well. I really hope their MacBU team does a good job with the next version of Office for Mac.
Could they have done better? Well maybe, I haven't used it yet so I don't know. I will say the default blue-ish color scheme is garish to say the least. However, it does look like Microsoft is finally starting to realize that adding cute animated characters to a poorly designed interface doesn't help. No one really wants to learn how to use a program, they want a program to work as they would expect it to.
What interests me more is how they were able to pull this off from the programming side of things. The Win32 event and windowing model does not lend itself well to these types of interfaces. It would take some smart developers to come up with the necessary framework to implement this type of interface. In contrast, Cocoa's target/action system makes writing these types of interfaces a breeze. The OpenOffice.org team has one hell of a task ahead of them now.
Still, OO.org hasn't been all bad. I think its biggest contribution to office software is hilighting how completely crappy Microsoft's Office interface has been and showing Microsoft that if they want to compete they are going to have to do better. My guess is that it's going to be several years before OO.org catches up unless they move to some sort of toolkit that does this stuff for them. If I were them, I'd ditch the crap they have now and move to GNUstep and put some development time on it. Of course, it would be blasphemy to use Objective-C at Sun since Java is supposedly the evolution of Objective-C. Having used both I can say it seems to me that Java is a step backwards not forwards. The reality here is that OO.org is basically dead because Microsoft just killed it the old fashioned way: by making a better product.
Jimmy Carter, is that you?
FYI, no one is going to consume less. Not going to happen. Ever.
However, if you want to take a first step I have a modest proposal: simply kill yourself. You will use much less energy that way. That is, after all, the logical extreme of reducing energy consumption.
I'd hire an electrician that I could tell was seeing the broader picture of the wiring system in the house, not just slinging wire through it.
Several companies do intra-company billing. Not all, but those that don't probably should. As shown by this discussion thread it's killing your business if you don't. Simply replace "that's not a supported configuration" with "this is a problem specific to your configuration. Do you have a charge number?" and things will move along much more smoothly.
Of course, some idiot managers will shit when they see how valuable your time is and swear up and down that they can get it cheaper elsewhere. And they can go right ahead and do that. Assuming you're running your IT department as a tight ship they'll come crawling back to you real soon. After, of course, blowing much of their budget on some consultants.
Ding ding ding ding ding. Give this man a prize!
Now, granted not all of IT is direct user help. Sometimes IT does for itself to make its job easier and sometimes IT works in general for the whole company to improve infrastructure. However, when IT is working to help a specific user or a specific department then it needs to be accounted for that way.
This is normally referred to as a "Service Center". And how do I know this stuff? Because I'm a developer and general IT worker (it's a small company) who does work on internal accounting software. I'm fortunate enough though to be in a place where our accountant organizes things as if we were a large company who wanted detail on every cost. This is particularly important because we are a mid-sized government contractor. When I work to improve infrastructure I work on one charge number. When I work helping some id10t user with their problem I work on their money; I don't mind, I'll take all the time they need to understand it-- within reason of course because there is still opportunity cost involved. I certainly NEVER refer to them as an idiot to their face or around the office (gossip spreads!!!) and I attempt to understand what they are doing wrong or what our systems could do better. When I develop software for accounting for costs on a specific contract I charge to that contract's overhead. If I write general accounting or business management software I have yet another charge number to apply it to G&A.
It's not hard at all for me to do this and it gives everyone the raw data they need to come up with an accurate idea of where money is going. It keeps our costs low and keeps our business efficient. I have to say if your business isn't doing it then you need to get a better accountant and maybe get or write better accounting software. If you're not doing real timesheets indicating how much you worked each job on each day then start.
Presumably, if you already have a copy center that functions this way, you probably already have at least some of the accounting infrastructure you need.
Let me guess... are you an IT manager?
I will give you that when you are working on your own overhead that it quickly becomes clear that supporting all sorts of different configurations takes a lot of resources. However, if you are able to charge to the user's job then it will quickly become apparent to the users (or at the very least their managers) that your time is valuable.
Unfortunately, many companies simply do not have their accounting and timekeeping systems set up to handle this situation. In effect they are letting their accounting software dictate how the business is run. It's a shame really but hey what are you going to do? You're IT. That's not your problem, right?
Thank you! I was about to make a similar post.
How is the imperial system is "handicapping" us? If anything, most educated Americans are able to work in units of either system. And as you pointed out, it's not as if we don't decimalize where appropriate. In addition to your machine shop example a more common example would be gun calibers. I'm sure everyone has heard of a .22 ("twenty-two") or a .357 Magnum ("three fifty-seven Magnum") or a Colt .45 ("Colt fourty-five") or a .50 caliber Desert Eagle ("Fifty caliber Desert Eagle"). All of those are expressed in inches. For reference, a 9 mm would be about a .354 caliber gun (simply convert 9 mm to inches).
All of those are in inches but nobody says a "point two two" or "point three five seven" or any weird crap like that. Which is, I think, yet another example of how the imperial system caters more to speech and common units. No one in their right mind says 2 feet 3 and 3/4 inches. If you really _are_ measuring down to 1/4 inch precision then it is 27 and 3/4 inches. Now say you are doing this in metric. Is it 70.5 centimeters? In that case, probably yes. But what if you're measuring 74 and 1/2 inches? Note that that is 6 feet 2 and 1/2 inches but no one would ever say that. Is that now 189.2 centimeters? Or is it 1.892 meters? From an american perspective the centimeter measurement makes more sense to me since that is the unit I'd most likely be measuring. But god knows what the crazy rest of the world does here.
I also want to reiterate your point about fractions. Halves and thirds and fourths and so on are extremely common divisions in everyday life. It's nice to know that a gallon has 4 quarts and a quart has 2 pints and that there are 2 cups in each of those pints. Nice simple division by powers of 2. Should make a lot of sense to a bunch of computer geeks wouldn't you think!? Interestingly enough, all of those can be represented without problem in a binarialized (is that a word?) system. Computers do in fact store floating point as ?/2, ?/4, ?/8, ?/16 and so on with the exception being most good desktop financial calculators which are explicitly designed in hardware to store in tenths and hundredths and so on to avoid rounding errors on financials which are now all decimalized. Oddly enough, the old pieces of 8 system made a hell of a lot more sense from the perspective of trying to store it as a floating point number in a computer.
My guess is that metric is going to take off in the U.S. about as well as Esperanto has. Languages and measurement systems are born out of the necessity for peoples to communicate, not through government mandate under some grand socialistic scheme.
Have you had a look at Apple's WebObjects?
It is a completely different beast from something like Hibernate or even Rails. Like other frameworks it maps database tables to entities (object classes) and database rows to instances of those classes and database columns to accessors in the class. And like other frameworks it is able to model relationships from one entity to another. And of course it does pull in data and cache it in application memory. Of course, any database access is bound to do this at some point.
However, unlike some persistence frameworks it does not attempt to enforce a hierarchical view of the data. In fact, it truly is a relational system. The big difference with WebObjects is the concept of the editing context (EOEditingContext class). The editing context is responsible for tracking all changes to the object graph. It is capable of working with hierarchies of course but it is also capable of a whole lot more.
Essentially, as the user modifies objects and modifies relationships between objects the editing context keeps track of what is going on. Ultimately, when saveChanges is called on the editing context (NOTE: NOT on some object) all of the changes are batched into the database within a transaction block. That may cause one object to be modified, another to be inserted, some to be deleted, or any combination of those.
In general, inserts are not going to conflict because one typically uses a database sequence to generate integer primary keys or is using compound PKs with enough unique user-entered information that they would not conflict. Should they conflict, EO is able to trap this. With updates, EO defaults to an opportunistic locking scheme whereby update statements are qualified with all attributes of the object (not just the primary key) such that if the record has been changed in the database the program can trap it. There is a facility for defining what merge behavior is appropriate for the particular entity by implementing it in a method that will be called in the event this occurs.
Also, unlike some other frameworks it is able to model EVERYTHING you can put in a SQL database. Or a non-SQL database for that matter! I have quite successfully used it to access a (copy of a) number of COBOL files as if each COBOL file was a table in a relational database. It helps of course that the COBOL data store was sort of relational to begin with of course. Where COBOL arrays were used I have successfully defined views in PostgreSQL to unwrap these into a set of related tables which I then model in EOModeler.
I'm not saying that the Enterprise Objects part of WebObjects is perfect, but it's as close as I've seen. The reason it is so good is that it truly follows the relational model and if you really think about it it sort of is an RDMS in and of itself. Think about the act of opening an editing context then either calling saveChanges() or revert(). Does this sound like "BEGIN" followed by "ABORT" or "COMMIT" to anyone? Of course, it goes beyond that since it's doing all of this in its own memory. Gets even better when you nest editing contexts. Suppose you have a deep hierarchy of objects. Yes, I realize we're dealing with a relational system, but sometimes you really do want hierarchy. In an app using EO the user is able to drill down from the root class and change all sorts of things about it and its children and its children's children and so on. Only when the user saves back up to the parent does it commit all of this in one big block. Primary keys and foreign keys are taken care of based on the rules you specify.
It really is a pretty damn amazing system. If you admire the relational data model I assure you you will like EO. Even better, model some DB you have and point D2W at it. It is absolutely amazing. Unfortunately the UI was designed by engineers in the early to mid 90s so it does tend to look like shit. But it WORKS. And looks are easy to change. Probably EOs biggest downfall is that it does so accuratel
My first experience with a computer was one that I programmed in BASIC at the age of 5. I'm sure there's plenty of other Slashdotters with a similar experience. It's not unreasonable to think that some children in a third-world country might have the talent for programming.
The only problem I see is that without access to decent tutorials and documentation it is hard to learn programming. Perhaps what's on the internet will be enough but perhaps not. In my case I had the Hands-on BASIC for the IBM PCjr (ISBN: 0070491690) book as my starting point. Wish I still had a copy of it. I remember it being really easy to read and follow and I was a child at the time.
I noticed that a comparison to Objective-C is quite conspicuously absent from the list of languages compared to D. Why is it missing? Granted D seems to be a much greater change to C than Objective-C is but I can't help but thinking that one of the main attractions to D seems to be its heap-based garbage-collected object system. You can already get the object runtime with Objective-C. If you use GNU you can even have Boehm GC (which is apparently the GC that D uses). If you use Apple you will have to wait for Leopard to get GC. Another new Objective-C feature is the ability to use full C++ objects as instance variables in your Objective-C classes and do the right thing with initializing (calling the default no-argument constructor upon alloc).
On top of that, Objective-C actually includes tons of reflection information. Although Objective-C has protocols which are roughly equivalent to Java/C# interfaces they are almost completely unnecessary. In Objective-C one can query at runtime whether a method is implemented or not and if so call it. So whereas in Java you'd do this:
in Objective-C you can do this:
The difference being that in the Java case you have to declare MyInterface as containing the one doSomething() method and inform java that your object extends MyInterface whereas in Objective-C you merely need to provide a doSomething method on your object.
Basically that means that in Objective-C every single method effectively becomes an interface. You would not believe how useful this is once you realize it. Note that at runtime there is ZERO difference. In both the Java and Objective-C cases the object is being checked to see if it implements something. Same with C++ if you use dynamic_cast<>()
Granted every language has its niche and I'm sure D will find its. Objective-C's niche is definitely GUI programming. The ample reflection information allows for easy implementations of archiving (serialization) and most importantly key-value coding and the related action methods pattern. It's a pretty damn cool thing when your RAD tool simply outputs archived objects that refer to methods to be called upon certain actions simply by name.
No, you credit a liability account. Regardless of the account type credits are always negative and debits are always positive. On a liability account a credit, being negative, increases liability. Unfortunately, under the traditional accounting system negatives are never used and thus it is recorded as positive credit because it is an increase. But it is an increase on an account which naturally has a credit balance and thus it is effectively negative.
Further confusion arises because when your bank prints your statement and talks about your account they refer to their liability to you. Thus, when you deposit a check they credit "your account" and when you withdraw money they debit "your account". In reality, it is not in fact your account but is rather their account of what they owe you if you decided to withdraw all of your money immediately.
Obviously they don't ever let you have a debit (note: positive) balance because then you'd owe them money. They always want to owe you money because that is the business they are in.
This is perhaps more readily apparent if you come up with a more similar transaction. Say you earn interest on money in your savings or checking account. You would of course expect them to credit "your account" for the amount of that interest. This is a good example vs. deposits/withdrawls because it's not a transaction you initiated. Now, say a company gives an employee some stock options. What they are doing is giving the employee money. So, wouldn't you expect them to credit their ledger?
Another way to think of it is that if you have a debit balance on your ledger (say you are a banker) then that means someone is in debt to you. Or you could say that if the banker has a debit balance on his ledger then you are in debt to the banker. The etymology of debit reflects this.
Now, let's get back to the stock options. Put on your corporate accountant hat. You (the company) has granted options to an employee. This means the employee has the option to buy up to x shares of stock at the price it was when you granted the options (or possibly when you declare the options have been granted). When at some unknown time in the future this transaction occurs the employee is going to give you cash. You know this amount in advance because you have fixed the price the employee will have to pay for the stock. What you don't know is what you'll have to pay (an expense) to buy this stock for the employee. What price will shareholders charge and thus what will your cost for that stock be? When the transaction ultimately occurs it is two transactions. One will take money from the employee and debit (increase the value of) some cash (equivalent) asset account. The other will cost you money and credit (decrease the value of) some cash (equivalent) asset account. Since the employee is unlikely to exercise options that cost him money you can be sure that ultimately it will cost you money resulting in a credit to your net worth. But how much and when? Who knows!
It is at this point where I have to leave it because I am not an accountant but rather a programmer. All I can tell you is that there are various accepted models for determining the price you will probably have to pay for the shares in the future. That is how you can determine your effective liability. It is very much a "guessing game" since the best you can do is use some probabilistic method.
This is all assuming of course that you do want to record the liability. There is also some chance that the options will expire and never be exercised! It's not so much that I'm not an accountant but more that I'm not a fortune teller.
Ultimately, a stock option grant results in an increase of projected liability. But in recent years some people want to see this accounted for not as an increase in projected liability but
Wow.. this is so wrong:
As other replies have mentioned, CDMA stands for Code Division Multiple Access. Each packet is tagged with a small code similar to the way ethernet packets are tagged with a MAC address. GSM is a TDMA system (Time Division ...) which relies on phones transmitting in an allotted time slot (milliseconds). This could be likened to a token ring network in some ways
That is bad enough but then you go on to state that it's not like a cable modem link. Huh? It is exactly like a cable modem link. In fact, just about any network or bus, wired or wireless, uses some sort of scheme to allow for multiple transmitters. In DOCSIS you have both a MAC (code) as well as an allotted time slot. DOCSIS takes it a bit further and does ranging. Basically, it pings the node at the head end until it gets an accurate picture of how long an electrical signal takes to reach the node. This allows the cable modem to adjust its transmissions such that when it transmits in a given timeslot it won't collide with another modem's time slot due to signal delay differences. The speed of electrons is after all only so fast.
I could go on and on here by pointing out that for example a T1 line is actually 24 time divided channels or that a PCI bus works on an acquisition basis. Next time before calling the parent poster a dorkwad (what are you 11?) you might actually want to know what you are talking about.
This is very good advice and I recommend everyone heed it when voting tomorrow. In general I tend to vote republican because I like their positions on issues. I am in favor of people getting paid what they will take, not a predetermined national minimum. I am in favor of health insurance not nationalized health care. I am in favor of giving parents the economic freedom they need to choose where they send their children to school, not forcing them to use a failing public school. I am in favor of allowing the executive branch to intercept communication with one endpoint outside the U.S. without a warrant. I am in favor of parents being notified if their daughter is going to have an abortion just as they would be notified with any other medical procedure. I believe a woman who shoots herself in the stomach two days before delivery should be charged with murder. I believe abortion mere days or even weeks before expected birth date is clearly murder.
On the other hand, I'm in favor of allowing gay couples to make contracts providing marriage-like benefits. I am in favor of legalizing abortion legislatively, not through some assinine court decision involving a legal house of cards. I would hope though that we could have some working sex education that explains to kids both the proper usage of condoms as well as pushes the idea that abstaining from sex is a wise decision for someone not prepared to be a father. Of course, that leads into a whole new political quagmire.
You see, I simply believe that a man should stand on his own but that a man's community should offer him help to get back on his feet when necessary. I do not believe in giving welfare checks to people for having babies. I do not believe in shoving the kids into public schools then lowering the testing standards so everyone can "pass".
I could go on and on but I'll stop here. What does that make me politically? I guess more a republican than a democrat. Although not entirely because there are a number of good democrats with views very similar to mine since these are I believe rather moderate views.
I am in Virginia's 2nd district. That means I get to vote Drake (R,I) vs. Kellam (D) and Allen (R,I) vs. Webb (D). Unfortunately, most of what I've seen from both sides is a negative campaign. I have seen and heard _some_ issues talked about by the democrats and some talked about by the republicans. And when I ignore all of the negative ads and focus on the issues ones I find I like the positions of the republican candidates more.
The thing is though, go to say Phil Kellam's website. Can you find where he stands on any issues? Hmm. Neither could I. His website has been this same thing primary attacking Drake and highlighting his great work banishing city stickers in Va. Beach. I will say his website looks "slick". Sort of like he does. Clinonesque I guess. The only place I can find anything is the "Media Center" section. Even there I find crap about Foley and the GOP did this and Drake did this and so on and so forth.
Drake's website isn't much better. The best section is again the "News" section which is showing news stories from local papers. Drake's site still has an awful lot about her challenger but has at least some about what drake stands for. Not as much as I'd like though
Do I like the way things are going in Washington? Not entirely. Am I going to go out and vote Democrat to send those damn Republicans a message? FUCK NO. What message would I be sending to washington? Run a better smear campaign and don't ever talk about the issues at all and you can win? Great, next election cycle we can expect more of this shit because it obviously worked!
How things will come out tomorrow is anybody's gues
Dom, I think you hit it spot on with respect to multiple monitors. I think you have it wrong about larger monitors though. Even on my 23" display the menubar is close enough (only need to bang the mouse up there with a quick flick of the wrist) that it's still highly usable. Not to mention that being the geek I am I tend to remember shortcuts for things I do often. Also, usage of the menubar can be reduced when drag and drop is available.
I think the reason for not having maximized windows is that it is a multitasking computer and that going in to single-task mode is highly ludicrous for your typical workflows. With that said, the iApps tend to be designed around the one maximized window model. It seems to work for those because they are like their own self-contained bottles (i.e. this is the microsoft model). I don't think that people really want to eliminate distractions. I think that some people are just neat freaks. Of course, using "Hide Others" from the application's menu (the one in bold) will handily eliminate all distractions. Assuming, of course, you don't have a billion icons on your desktop.
With spaces in leopard eliminating distractions will be even easier because one can simply configure a space with only the things they're working on. This, I think, is Apple's answer to the good things about maximized windows. That is, that they eliminate background distractions.
The best way I can explain it is that the menubar on OS X is not much at all like the menubar on other systems but much more like the contextual menu (i.e. right-click menu) that other systems have. It follows what you are working on. As you change apps all of the top-level labels change. As you move around between things within an app items in the menus are disabled or enabled as needed. Of course, OS X also has contextual menus but because Apple has a full menubar with all of the commands in it the contextual menus tend to only have a few well used ones.
If you stop thinking about the menubar as controlling the window and start thinking about it as controlling the specific thing you're working on then it makes a lot more sense. It's funny you should bring up The GIMP since it actually uses the right-click on image gets a menu model. And it's not your typical contextual menu but rather a replacement for a menu bar. That model is almost exactly the Mac model except the menu moves to the top of the screen horizontally and all the top level labels are visible without having to first right-click. Apply that to the entire OS and maybe you can see why it's not such a bad idea after all.
Of course, it does have its problems. If you twitch-click on your desktop or inadvertently miss the menubar by a few pixels you will change the menu to the Finder's menu. To me the simple solution here is to get rid of the desktop entirely. NeXT for instance did this. NeXT didn't have a common menubar but clicking (left mouse button actually) on the desktop brought up the menu which is essentially a GIMP style menu (or is it the other way around?) for the currently active window.
The desktop metaphor sort of made sense in the early Mac days when it was showing basically what was on the currently inserted floppy disk. You double clicked something and it came back to life as a document. Now that we have much more storage space the whole idea of the desktop is broken. I am all for the NeXT model of making the desktop into a workspace where only windows exist and making the file browser just a normal app.
I wish I had mod points. Very well done.
I believe it's actually a design decision on the part of Apple. The traditional way to move or copy files on a Mac has always been to use the mouse to drag them. This isn't hard at all when you have a decent sized screen and you can simply stagger the source and destination windows then drag from one to the other.
It is interesting though because dragging files is really something someone needs to be shown. My experience has been that people don't just pick it up without at least some minor prompting. Once you show someone on a Mac they seem to understand it quickly. However, I've had a hell of a time showing how to do it on Windows PCs. It just seems that people can't get their mind out of the one maximized window mindset and it's rather hard to drag from one maximized thing to another. Of course, you can drag through the task bar but that's another learned behavior, one that doesn't make that much sense compared with a normal drag.
This, I think, is one of the major shortcomings of Windows. Microsoft has basically crippled the UI to the point where it's nearly impossible to run more than a few apps with more than a few windows open. Unfortunately, it seems that Vista doesn't really fix this shortcoming. They have a cool looking alt-tab replacement but it's just that, cool looking.
It would be very hard for Microsoft to move to the Mac model here. Part of the Mac model is that the menubar switches with the app you're using and that all the toolbars and pallets disappear when the app is not active and switch when you switch which document you're working on within an app. Contrast this with the Microsoft style of putting giant sidebars on all four sides of the document area within the window. It makes the windows too big to be sized anything other than maximized on many screens.
Of course, some people have a preference for the Windows way. They say it "looks cleaner" because they only see what they're working on. Maybe some people really get distracted by having portions of other windows behind their active one still visible. Funny enough, that aspect of OS X never bothered me. I found it relatively easy to get used to the idea that windows generally exist on the screen and don't try to own the entire screen. To me it seems similar to the way one stack of paper sometimes obscures another on my real desk. I never stack everything neatly in piles and grid them out like tiles. I've got one pile of papers that's half covering another so I can see at least part of what's under it to know it's there. This way I can put a lot more crap on my desk and still know where it is. Now I know I'm not the only person whose desk looks this way
Still, can I really blame Microsoft for these things though? Not really. They made these decisions years ago trying to get people to move from DOS to Windows and then later from Windows to newer versions of Windows. The latest trend I'm seeing is for some people to get dual monitors on Windows. This way they can have two apps maximized, one maximized on each screen. I ran dual monitors on OS X for a while but lack of real maximization (and no desire to have it either) means you wind up with a good sized worksurface with a huge line in the middle of it. I've since decided that Apple is defiitely on the right track with the bigger displays. Particularly if you have the 23" you can begin to see how it completely changes how you want to interact with the computer. You're not going to maximize things; even at the smaller 20" size a window would be ridiculously big. What I find myself doing is just staggering more and more windows all over the place. It looks just as messy as my real desk. This, I think, is exactly the point. Apple has taken the desktop metaphor one step further with these huge displays.
And what has Microsoft offered us? More of the same. Compu
It doesn't have to be this way. Take Objective-C for an example. All messages are identified by a selector which is actually just a plain old C string. Take for example [NSView -frame] which returns an NSRect giving the frame (window) size and location. FYI the syntax in objective-C to call it is NSRect frameRect = [someView frame]; If it were more C/Java like syntax it'd be NSRect frameRect = someView->frame();.
Anyway, getting back to the point: The way the compiler compiles this is a call to objc_msgSend(someView, ___selector_for_frame___) where ___selector_for_frame__ is simply a C string. Recall that C strings are simply pointers to an array of char. What the compiler does is unique all references within a single compilation unit (a .o file) such that all uses use the same pointer. Furthermore, the link editor (ld) sets things up such that they all point to the same thing within the executable or library or framework you are linking. Then, the dynamic linker ensures this when it loads in dynamic code. This way, the runtime need only do a comparison of the pointer itself (where in memory it is pointing) rather than checking what it's actually pointing to.
Basically, that is to say that the test someSelector == someOtherSelector actually works. It is not necessary for the runtime to do strcmp(someSelector, someOtherSelector) == 0.
I assume similar tricks are used in other dynamic languages. I only mention Objective-C because it seems to be the one with a bunch of publicly available documentation on its inner workings.
That's what the tabs are for on the undersides. To me it just came naturally to put the tip of my thumb on the left tab and the tip of my ring finger on the right while holding the button down with both my index and middle finger.
The mouse is light enough it hardly takes any effort to do this at all.
I have larger hands than most people (though not compared to my size since I am 6'6"). The Apple mouse has been great for me. The thing about the Apple mouse is that you don't grip it like other mice. I don't (can't actually) move the mouse around on the pad with my hand squarely on top of it. What I do is rest the palm of my hand (except the heel) on the top of the mouse. I let my thumb and pinky rest on the mouse pad. My 3 middle fingers curl over the top of the mouse. The heel of my hand naturally curls over the back.
Basically, it is very much like grabbing a ball. The other thing is that you will notice Apple's accelleration curves are fairly slow. Because of this, you tend to need to pick up the mouse from time to time to do a drag across the screen. The way the Apple mouse is shaped it's very easy to do this since you just sort of squeeze it. Unfortunately, some idiot at Apple decided to make that an action so every so often I inadvertently activate Exposé.
Other than that, the Mighty Mouse and the Pro Mouse before it have been great.