Thursday, January 29, 2009

Ctrl-backspace is broken

Something that continues to really bug me about using Windows is the interface is just so inconsistent. And I'm not talking about Microsoft-delivered apps look and act differently from third-party apps - I mean different Microsoft apps just don't behave the same.

Case in point: Ctrl-Backspace. Why doesn't it do the same thing everywhere?

Under Linux, Ctrl-Backspace always did the same thing, no matter what application I was using: it deleted the current word. For example: I always have a long passphrase, usually 10-12 characters long. When typing my passphrase under Linux, if I got 8 or 9 characters in, then felt I'd mis-typed a key, I'd just hit Ctrl-Backspace and start over.

Under Windows, this doesn't work! At login (or when unlocking the screen) Ctrl-Backspace inserts a Ctrl-Backspace character. How confusing!

Yet in other Microsoft apps, Ctrl-Backspace works "properly", and deletes the current word. It does that in MS Word, MS Powerpoint (but not MS Excel), and Internet Explorer. Use Notepad, and it inserts a Ctrl-Backspace character - but Wordpad works "properly".

I even found another instance where Ctrl-Backspace just backs up over the last character (just the same as hitting Backspace) and another instance where Ctrl-Backspace does nothing (I'm looking at you, Excel.)

These are all Microsoft apps, but they don't act the same. Why the inconsistent behavior!?

I hate Office

Microsoft Office 2007 has to be one of the worst application designs I have ever seen. I just lost an important document I was writing, because the interface is too broken.

I mentioned before that the MS Office logo in the upper-left corner is actually a menu. In previous versions of Office (and any other Windows application - or Linux/GNOME application, for that matter) the program icon in the upper-left has basic window operations: minimize, maximize, etc.

But Microsoft, for reasons known only to them, decided that MS Office would be different. The program icon is actually a menu, where you can open, save, print documents. And did you know that if you double-click on that program icon, MS Office will close automatically?

I didn't know that ... but I just learned it the hard way. Accidentally clicked on the program icon, clicked it again to make the stupid menu go away. And there went 20 minutes' work on an important document. Now I get to write the whole document again. Gee - thanks, Microsoft.

Wednesday, January 28, 2009

Paint doesn't have basic functionality

I remember back in the early 1990's when this "Windows" thing was still a cool, new platform that everyone was experimenting with. One program that was considered generally useful was MS Paint, which allowed you to create simple drawings.

MS Paint hasn't progressed one bit since then. Actually, to give Microsoft some credit here - looking at Wikipedia, MS Paint hasn't progressed since 1998.

Under Linux, I often used GIMP for simple editing of pictures - photos, screenshots, etc. I certainly don't consider myself a GIMP expert, but I used GIMP once in a while to touch up a photo, create a graphic for my web site, or to highlight things in, say, a screenshot. You can do some pretty advanced stuff in the GIMP, but I never did much beyond layers and a few simple "canned" effects.

As an example: I once needed to give campus directions to a friend visiting from out of town. Google Maps didn't have campus walking directions, so I opened the university campus map, in GIMP. Cropping the image to show just the part of campus my friend would be visiting, I used the Paths tool to highlight the route my friend would take get to his destination, including passing through and around a few buildings. Using the GIMP was fast and easy.

MS Paint, on the other hand, is way too limited: you can hand-draw some lines, or add cheesy spraypaint effects that might have been "neat" in 1990. I can't even crop an image! That's a basic operation, and I can't do it. I guess the only way to "crop" in MS Paint is to make a selection, copy, then paste. But that replaces the image you're working on - you cannot have more than one image open in MS Paint, apparently. Lame.

I don't mean to compare a jackhammer (GIMP) with your basic sledgehammer (Paint.) I know that MS Paint is simple, not really meant for anything more than basic operations. Clearly, Microsoft assumes that someone who wants to do more than just basic image operations would pay $$ for a "professional" graphics tool like Photoshop.

But hey, the tool (GIMP) that is provided by default and for free with any Linux system is way more powerful than the tool (Paint) that's provided with Windows. I'm comparing default-tool to default-tool. Hands down, Windows loses!

PrtScr is retarded

I was going to write a different blog post today, about the inconsistencies across the user interfaces in Windows. I'll write that one later, maybe tomorrow. But to demonstrate the inconsistencies, I wanted to take a screenshot of some programs. That's when I learned a terrible lesson in Windows:

PrtScr is retarded.

Under Linux, when I hit PrtScr, I was immediately prompted to save the screenshot to a file. Very easy, very immediate. I knew that PrtScr was doing something.

In Windows, I figured it would work the same way. So I hit PrtScr - but nothing happened. Didn't matter how many times I tried PrtScr, I got nothing. It wasn't until I did a Google search for Windows XP PrtScr that I realized "print screen" really just saved a copy to the clipboard.

To me, this is just backwards. If I want to take screenshots of different applications under Windows, I have to do them one at a time, then paste them into (for example) Paint. That's just to save the screenshot to a file: PrtScr, open Paint (because I didn't want Paint in my screenshot), paste, save, close Paint. Under Linux, this process was easy: PrtScr, save to file.

Why does Windows make it so hard? I'd rather just hit PrtScr, immediately save the screenshot to a file, and move on.

At least Windows does it the Linux way and allows Alt-PrtScr to copy a screenshot of just the open window. But still no way to directly save a screenshot to a file.

Tuesday, January 27, 2009

One thing I like

I use a laptop, but I use my laptop at my desk for most (if not all) of my work day. If a laptop battery stays plugged in all the time, never letting the battery discharge, the battery doesn't hold much of a charge after about a year.

So when I'm at work, I sometimes like to let the laptop run on battery power, and will unplug the laptop from the wall.

Occasionally, I'd forget I'd done this, and would come back from a meeting to find my Linux laptop had automatically shut down when the battery reached a critical level. The default in Linux (Fedora) was to shut down. Plugged in the laptop, booted up, and Firefox restored all previous tabs and sessions, and OpenOffice automatically recovered any open documents.

This happened again the other day with my Windows laptop. I came back from a meeting, and realized my laptop was off. Except it wasn't - it was in standby mode. Plugged in the laptop, hit the power - and the laptop woke up. No browser sessions to restart, no Office documents to worry about.

So I guess that's one thing where the default on Windows makes more sense than the default on Linux. I'll give it that.

Wednesday, January 21, 2009


The font rendering under Windows is killing me. Or making me go blind. Possibly both.

Under Linux, I had 4 methods of font rendering to choose from: "Monochrome", "Best Shape", "Best Contrast", "Subpixel smoothing (LCDs)". I think I used subpixel smoothing, which gave me great-looking fonts that were very readable on my flat-panel display. I had installed Microsoft's core fonts, and usually wrote documents using Times New Roman or Arial.

Now, on the same Dell 1905FP flat-panel display, those same fonts are giving me a headache.

Why is Microsoft so limited with fonts? Under Windows, there are only two methods to smooth fonts: Standard (not smoothed, hard to read) and ClearType (blurry but smooth, and hard to read.)

After writing a Word document, or browsing the web for any amount of time, I get a headache from squinting at the screen. Sure, I've increased my zoom level in Word, and increased the minimum font size in Firefox, but the fonts look out of focus and are just plain hard to work with.

Monday, January 19, 2009

Diagrams and Word

I needed to create an overview document for our Director, and the best way to (briefly) provide an overview of our architecture was to draw a little diagram. This was never a hard thing to do in OpenOffice; just insert a new OpenOffice Drawing. OpenOffice also let me set up guides and snaps, so that all my diagrams looked clean.

My first challenge to do this in Word was figuring out how to create a new drawing. Under the "Insert" tab, there's a "Shapes" pull-down. Nope, that draws a shape directly into the document, hiding everything underneath it. "Picture" allows me to insert a picture from a file (for example, a JPG photo) which is not what I needed. "Word Art" wasn't helpful, either.

Eventually, I stumbled upon "New Drawing Canvas" under "Shapes". This wasn't intuitive to me. Everything else in the "Shapes" pull-down is a shape, except the "New Drawing Canvas", which is (technically) an action.

(Ironically, in writing this blog entry, I opened an empty Word document and tried to retrace my steps so I could write it all down. I honestly couldn't figure it out again, and I had just drawn 3 diagrams in my document for work.)

Once I had my new canvas, my big problem was that everything just looked sloppy. If Word understands the concept of "snap points" they are spaced too close together to be useful. Even in the absence of snaps, there are no guide lines to help me line things up by sight. I fell back to the old trick of "create a shape I want to duplicate several times, then copy/paste." It worked, but doesn't look nearly as nice as diagrams created using OpenOffice.

There isn't an "Options" for the canvas, but you can right-click and select "Format Drawing Canvas" to get some options. There isn't much to choose from, and I found no options that let me set up guides or snaps.

To create a simple diagram in Word, I had to fight the whole way - and I'm still not happy with the result. If someone can point me the right way, please do so in the comments.

Saturday, January 17, 2009

MS Office vs OpenOffice

I was at a conference yesterday, but I still needed to write some documents. I figured it was a good first exposure to using Microsoft Office 2007.

On Linux, I used OpenOffice (version 2.4) to edit all my documents. Usually Word (DOC) documents, but the also some Excel (XLS) spreadsheets. Exchanging files with Windows users was not a problem, because we (mostly) used the same DOC and XLS formats. A few people in the office use the newer DOCX and XLSX file formats, but I had downloaded Novell's ODF Converter so was able to convert those file formats to versions that I could read.

Using the standard DOC and XLS formats meant it was easy enough for me to pick up right away with MS Office 2007 to edit my documents. Back to work, right?

MS Office is a pain. It's not the worst word processor in the world, but I find that Word actually makes it more difficult for me to create and edit my documents. It's much easier to do the same work under OpenOffice.

Maybe it's that I'm just new to MS Office, too used to how it was done under OpenOffice. I'm willing to give it a fair shot. Here are the main problems I'm having:

The menu works differently from every other Windows program.
I started by opening an existing copy of a document, but wanted to save it under a new name. It was not immediately obvious to me that this "file" command didn't exist under a "File" menu. In fact, there is no "File" menu - instead, you click on the MS Office logo in the upper-left corner. I had always understood that the program's logo in the title bar was where you could select "Maximize" or "Close", not the place where you choose "Save As". This makes no sense to me.

The menus are too weird.
Those "menus" at the top of the screen in Office aren't really menus - they're tabs. I am constantly confused by this. Again, no other Windows program works like this.

Adding and editing a page header is hard.
Most documents that I create for work require that I insert a standard "OIT" header. The process to do this was fairly straightforward under OpenOffice: Insert-Header, then paste the "OIT" banner into the header. Under MS Office, it's just hard. Sure, the "Insert" tab, then click Header-Blank wasn't too difficult to figure out. But once I had the header, the "Type Text" message really did not look like a text input field. It looks more like an information box. I clicked on it, nothing happened. Double-click, nothing. Why does this have to look different from other Windows applications?

The Style Gallery is confusing.
I like the concept of the Style Gallery, and the fact that you can preview a style as it would be applied to your text is a nice feature. But the rest of the Style Gallery feels poorly executed. At first, I thought there were only 6 different styles to choose from - there's nothing to indicate there are more styles, since the scroll bar is nonexistent here.

Also, why mix block styles with character styles in the Gallery? It means the "Title" style acts differently when applied to a paragraph than when applied to a selection of text within the paragraph. That doesn't make any sense.

In the end, I was able to create my document, but felt frustrated all the way. We'll see if this improves with time.

Friday, January 16, 2009

Day 1

Let me first provide a little background: I've been running Linux (mostly Red Hat, then Fedora) at work since 2002. The latest distro that I ran on my Dell D420 was Fedora 9, running GNOME. I'm a manager, so pretty much all I used was Firefox for web, Evolution for email, OpenOffice for writing and editing documents. (Although, I'd recently been using our work webmail system to read my email on the web - more convenient when I'm away from the office.)

Since I also work on a personal web site in my spare time, I made regular use of the GIMP (for updating the graphics) and Terminal so I could ssh into the web server.

Hardware: I'm plugged into a standard Dell D-series mini-dock, which provides me with optical media (CD-R, DVD) which is presented to the system as a USB optical drive. There's also a handy set of USB ports, which I have connected to my Dell 1280x1024 monitor. The monitor has USB ports on the side, which are easier to reach if you want to plug in a USB memory stick, and is also where I connect my USB keyboard and mouse.

On Thursday, January 15, the desktop support guys dropped off my "new" D430 laptop with the standard Windows image. The CPU is a little faster than the D420, but otherwise it has the same memory, wireless, etc.

From the start, it was a hassle. Where Linux easily recognized the mini-dock as a USB hub that happened to have an optical drive attached to it, Windows did not. After 15 minutes of messing with it, and a reboot, we finally managed to convince Windows that the dock existed. So now I had an optical drive. More importantly, I could use my external USB mouse and keyboard, rather than leaning way over to type on the laptop while it was still in the dock.

Next, we fought with Windows to get it to recognize my 1280x1024 external monitor, which Windows was convinced was 1024x768. The support guy recommended rebooting with the laptop lid closed, thinking that Windows didn't like driving two displays at different resolutions - we tried that, but rebooting didn't fix the problem. The support guy suggested the problem was with the mini-dock, but I didn't buy it. Eventually, we figured it out (I don't remember what he did) but it required yet another reboot for the resolution to show up properly as 1280x1024 on my external monitor.

For those of you keeping track at home, that's 3 reboots to get basic functionality working on my laptop. Not a great first experience with Windows. I'm off to a great start.

Linux in Exile

I've been a Linux user since 1993, when I was a student at university. Until 1998, I had a partition on my home system to run some Windows games, and since 2002 I've been fortunate enough to run Linux full-time at work. It has been a great experience so far.

That all changed this week.

For reasons I won't go into, I've been asked to move back to Windows, at least for work. The difference between Windows and Linux has been shocking, to say the least. Since I find it interesting when long-time Windows users experiment with Linux for the first time, I thought it might be equally interesting for this long-time Linux user to blog about my first experience running Windows in over 6 or 7 years.