Recently, getting answers to questions seems to have come a little more difficult for many new member, so Access Page Editor JDettman took on the task of coming up with some tips.

While EE has a huge searchable database of Previously Asked Questions (PAQs), which often yields a solution to a problem, there may be times when you need to ask a question to get a solution. What happens when you ask a question?

First you'll need to find a Topic Area (TA) in which to place your question. Many TAs have a specific focus, such as the Microsoft Access TA, which deals strictly with questions pertaining to the relational database product produced by Microsoft. Other TAs have a broader scope, such as Routers, which covers every kind of router ever made. There are some general "catch all" TAs as well. For example, from the EE home page, if you click on the "DB" tab at the top, you'll find yourself in the TA for databases in general. But if you look off to the left, you'll see additional TAs for specific databases such as Microsoft Access, Visual Fox Pro, SQL Server, and so on.

You should do your best to find the correct TA to place your question, as this will get things started off on the right foot. Placing it in the wrong TA is not a major problem; it just may delay you in getting an answer.

Once you've located the proper TA, click the "Ask A Question" link. You should first verify that you are in the correct TA, which is displayed in Step 1. You'll then need to provide a title (Step 2), a description of the problem or the actual question (Step 3), and a number of points that you feel the question is worth (Step 4). The title is what appears in the TAs lists of questions along with the date and the number of points you offered. It is important to pick a title that accurately describes your question. Something like "URGENT HELP NEEDED" really doesn't help all that much.

TIP: Typing everything in upper case makes things difficult to read. You should always type in mixed (normal or proper) case.

Every Expert will now need to look at the question to figure out if they have any knowledge of the problem, which wastes time. If everyone did this, we'd get nowhere fast! It doesn't help either when the question goes into the PAQ database as it makes searching difficult. So try and make it something descriptive: "Have database corruption - Urgent help needed".

TIP: Only the first 40 characters of the question title appear when listed in the TA, so make those characters count!

With the title out of the way, the next thing is the question itself. One of the first steps into getting a good solution is in asking a "good" question. Consider the following: "I need help with my Access database!"

Just like the title before, this doesn't really help all that much. So what goes into asking a "good" question? Well keep in mind the following:

1. The Experts don't know what your exact situation is.
2. The Experts don't know your skill or knowledge level.
3. All the Experts are volunteers and will do their best to help you.

What you need to do is provide enough detail to explain you problem in basic terms, but not so much detail that it makes it difficult to understand what the question is. It takes asking a few questions before you get the hang of it, but as you ask more and more questions, you'll get better at it! If you're not sure about putting something into a question, then don't. It's better to be brief and let the Expert(s) question you about your situation so they have a clear understanding about what is being asked rather than clouding things with a lot of useless details.

That's the second part of getting a good answer; everyone has a clear understanding of what exactly is being asked. Make sure the Experts understand what your asking and all the factors involved. It's also important that you don't change the question midway through. For example, I've often seen questions where after a few exchanges the member says: "But what I'm really trying to do is this..." You should talk about exactly what it is your trying to accomplish.

The third part to asking a good question and getting a good solution is to keep the discussion focused. Don't ask additional questions unless it pertains directly to the original question.

The last thing you need to do when submitting your question is assign a point level. Simple questions where you expect a one or two line answer should get 20 or 50 points. Questions that will take a little more effort and may involve several exchanges should be in the 100-300 point range. Complex questions should receive 300 points or more. The maximum amount of points that you can assign to a question is 500.

After you've submitted your question, you may have one or more Experts respond. It's important to keep working with the Experts until you get an answer to your problem. Sometimes the answer to a problem is that there is none and you simply can't do what you want. The critical thing is to not give up on a question. If you're not happy with the responses you're getting or don't understand what is being said, just say so.

Once you and the Expert(s) have come to the end, it's important to close the question and assign a letter grade. Note that you can also split the points between Experts if you feel more than one contributed to the solution. When assigning a grade, keep in mind that the grades of A-B-C are not the same as the grading in school. An "A" grade, in general, means the Experts helped you by giving a complete solution to your question, while a "C" grade means that the comments by Experts did not completely address your question, but may have helped you along.

TIP: Before assigning a lower grade it is always a good idea to give a quick explanation of what you feel is incomplete or missing from the solution, thereby giving the Experts a chance to follow-up with more complete solutions. Often it is simply a case of misunderstanding -- where the Experts thought they gave you enough information, but you needed more details.

What if you find the answer yourself after submitting the question? Post a comment to that affect along with the solution you used. You can then request a refund of the question points in Community Support.

For additional tips on questions (and the site in general), take a look at the Help pages or you can also post a question in the Community Support TA if you need some help in working with questions. You can also e-mail the Page Editor for the TA where you question resides. The Page Editor can be reached at [username] "at" experts-exchange.com.

What your responsibilities are:
  >  Try your best to ask a "good" question.
  >  Keep communicating with the Experts.
  >  Close your questions as quickly as possible once a solution has been reached.

What you can expect from an Expert
  >  Their best effort to help you solve your problem
  >  A professional and courteous attitude

Some Do's and Don'ts:
  >  Don't ask questions that violate the member agreement. Experts are not here to do your homework, help you hack into systems, or crack passwords.
  >  Don't accept an answer until your satisfied. Keep in mind though that in some cases there may not be an answer to a problem that you are happy about, but it is the correct answer.
  >  Do say so if you're not happy or don't understand something. Two-way communication is key to the getting a good answer or solution.
  >  Don't expect Experts to do all the work, like writing an entire program or app. They are here to help when you get stuck. If you do get stuck, be prepared to show what steps you have taken so far and explain where you are stuck.
  >  Don't ask for solutions to be emailed to you or take the discussion off-line as this violates the member agreement.

fritz_the_blank, the Page Editor for Active Server Pages, was at one time the top Expert in two of EE's busiest TAs: ASP and javascript.

Even though the use of session variables can be resource costly, sometimes the application logic makes them the most attractive solution. For example, if you have a series of forms that need to be populated, and you don't want to commit the data to the database until all of the forms are completed, the use of session variables to store the form values temporarily can be a very attractive approach: it allows the user to navigate both backward and forward without losing entered data.

In order to minimize resource drain, however, it is useful to destroy unwanted session variables when you are done with them. Of course, you can call Session.Abandon() to clear all variables, but what if you only want to remove certain variables rather than all of them? Moreover, what if you want to do so programmatically by looping through the collection?

At first glance, this does not seem to be an issue. Consider the code below. We start by initializing all of our variables. Next, we display them using a simple for loop.

Now imagine that we want to preserve for other pages in the site all variables that have the string "Name" in them, i.e., strFirstName, strMiddleName, strLastName, strNickName but to remove all other session variables. At first, one might expect that the code shown below should do the trick. However, when you test the page, we can see that is not the case!

What happened? Why have the variables for the state, web site, and fax number remained?

The problem lies in the interaction of two processes that don't work well together. On the one hand, we are using a for loop that iterates through the collection, and on the other, we are using the .Remove() method which modifies the collection during this iteration. As a result, the for loop will operate in an unexpected manner.

Fortunately, a solution is at hand. The first step is to loop through the collection to create a list of items to be removed. Next, we create an array based on that loop. Finally, we iterate through the array and remove each item in the array from the session contents as below:

Now when we test the page, we get the results we expect.

This article is significant in two respects. I have received many inquiries from programmers attempting to do something similar to the above with session variables, and they were convinced that the problem resulted from a bug in VBScript or with ASP. However, as I show here, this is standard for VBScript collections, and rather than being a bug, it is an error in program logic that is devilishly difficult to isolate. In a larger context, the same logic applies to dealing with VBScript collections in general, and the approach discussed here can be used in all such instances.

leew is one of our Windows Page Editors.

Beginning of the end for Microsoft?

In recent discussions I've had with other IT professionals, I've been postulating that should Apple decide to open up MacOS X for x86, Microsoft might well start a slow but rapidly growing downward spiral from its purch high above the desktop OS market. While I haven't reviewed all the details - indeed, I'm not sure all the details have been made public - it would seem Apple has begun the rebellion against the empire.

For most home users, the Mac OS has long been considered an easy system to learn and those who try it once often recognize it as such. With the release of OS X a few years back, Apple finally brought true multitasking and reliability to its platform and even I - a long time Windows geek - recognized it as a true, capable OS.

Now, if we stop and think about things for a moment, what might happen next? That all depends on Apple. But even if they do lock themselves into their own hardware, expanding beyond it is the next logical step. If Apple starts selling Macintosh systems based on x86 architecture, this could lead quite quickly to REAL choice in the OS Marketplace. Linux is fantastic, but still way to complex for the average user. Think about this for a moment - if Apple runs on x86 and rethinks the importance of controlling the hardware OS X can run on, you might be ordering your next Dell with Tiger on it. And imagine if Joe User can go to the local computer store and buy OS X competitive upgrade to be installed over Windows?

Of course, Microsoft wouldn't sit idle while all this went on, but in the coming years, price, innovation, and compatibility all might benefit from this week's announcement.

cracky is the Page Editor for all things Macintosh.

Apple *Hearts* Intel

Big news it seems. Apple has confirmed rumours that it will start making the migration to Intel chips as soon as next year. This seems to have caused a bit more of a ruckus than I would have expected. Although it was big news for me too, I was hardly in shock at the announcement. So, what does it mean for us? Do we grind everything to a halt in anticipation of a new wave of hardware and software from Apple and its vendors?

The first question people have asked is why? It may seem trivial to some, but closer analysis will tell us that the move was destined to happen. The driving factors seem to be in reliability of supply and the future of the chipsets. Due to complicated cooling mechanisms, a G5 laptop looks less likely as every day passes. Considering the exponential consumer trend towards mobile computing, Apples need to preserve its future in this market. Apple experiences higher laptop growth than many other vendors. Meanwhile, IBM is understandably shifting its efforts to supplying its powerful console partners with Cell processors for the Xbox 360 and PS3. Being a poor cousin wouldn't affect Apple much in the Intel camp, since it can piggyback on computer-specific advances, whereas the differences between Apple's requirements and those of consoles could very well hurt any future innovations at IBM.

Macs have a tendency to retain their value and continue to run OS upgrades better than their PC counterparts, so this gives some credence to those who fear the "Osborne Effect" on Apple's sales until the new machines arrive. There is a distinct possibility that consumers will hold off buying new gear until the updated hardware comes in. However, one would have to say that there's clearly no better time for Apple to make the change than now. Sales are rising, the iPod is an awesome crutch and they are quite the media darlings at the moment. Why wait for a flat spot when you can charge with a full clip of ammo, especially considering Redmond's track record with promises. Longhorn will likely be delayed, giving Apple ample opportunity to pip them at the post both with new hardware and Leopard supporting it to take line honours in the PR race. What happens after that is another matter.

The change has prompted questions about developers being left in the dark with their PPC code, but Apple insists that the underlying Rosetta emulator will give high performance not usually associated with emulation. If we are to believe them, we'd have to think that an early announcement would give developers plenty of opportunity to port their apps while avoiding any deadline jitters in the transition. Still others conjecture that the underlying x86 architecture gives rise to the possibility of a dual-boot system, which would throw some very interesting possibilities into the consumer market. This could see an unholy matrimony between Windows and Macintosh on hardware systems, with users taking greater comfort in a machine that will not require to play with all their toys in the one sandpit. Gaming and accounting software come distinctly to mind. It's here that Apple's attractive and reliable hardware could potentially steal sales from a PC power user base.

Apple looks confident to weather the consumer storm in the lead-up to the transition. Let's hope consumer confidence matches Apple's.

Further (lengthy) reading: The Register, Daring Fireball, and, of course, Apple.