Logikal Blog

Thoughts and words from Roland Hughes

March 23, 2015

KWrite and Kate

One thing which always annoys me about being forced to use a Gicky-Gnasty-Gnome based desktop is the total lack of indispensable packages. While, in theory, you can install KWrite and Kate under Gnome, they don’t run correctly. At least, they don’t run correctly on YAU (Yet Another Ubuntu) releases. I believe this is due in large part to Canonical not testing desktops together. They ship different products for each desktop so not much cross desktop testing needed in their view.

I spent a good many years in RPM hell with SuSE. I endured the ugly thing called YAST. I spent many an hour trying to recover from RPM installs which gagged part way through leaving the package manager in a state where one could neither uninstall the broken package nor repair it. Admittedly I was on dial-up and the package manager was absolutely worthless when it came to handling a lost Internet connection.

The thing which kept me using SuSE was their cross desktop testing. Back then when you installed SuSE you could install all of the desktops during the initial installation process. There was a high level of cross desktop testing. You changed your desktop with each login. Despite all of the other issues, I never noticed a problem where a KDE software package ran badly under Gnome or the other desktops.

Yes, when I’m doing development on Linux I’m typically using QtCreator because I’m developing with C++ and Qt, usually for some embedded target. That doesn’t mean I don’t use other editors. Here is what I tell the young developers on any of these projects.

Kate is your mommy. Tell your mommy everything and she will tell it to you when you need it again.

People laugh at that, and that is the point. You remember something which is funny. When they ask I tell them that Kate is more than just a fantastic editor. It’s a limitless shelf to store information you really need on. Once configured to automatically save the session on exit, you can have as many cheat sheet files as you want open in it. You click the one icon you add to the bar or desktop and one thing with all of your cheat sheets opens up.

A couple of projects ago everybody laughed at the Kate line, but by the time the project was over most of them had came to get one of my cheat sheets. Yes, they had theirs, somewhere, but mine were always right there. I didn’t have to remember file names or where I put that piece of paper.

KWrite has become another indispensable tool. Not so much for development. There have been a good number of years where WordPress ships an editor with severe gastrointestinal disorders. Not only is it slow and cumbersome to use, pasting to it from a text editor is fraught with peril. There was a brief period where Blogilo worked, then it was over. Just as brief a period where OpenOffice (before it was consumed by a corporation and dropped from most distros) where I could reliably past from OO. Once that stopped working I had to search for a text editor then reformat all of my embedded links. For a while the only text editor I could achieve some semblance of acceptable results with was Tea, then that quit working. Finally KWrite started working with WordPress. Yes, I still had to re-work all of my embedded links, but there was a 50/50 chance when I told the WordPress editor to perform full justification on the text it would actually work and that was the highest success rate I had found.

Recently I tested LibreOffice with WordPress and, in fact, I’m writing this post using LibreOffice. There seems to be some level of cooperation between the two packages now. That said, I still use KWrite a lot. I used it half a dozen times this morning while writing email just to look up spelling. Yes, Thunderbird flags the words as spelling errors and has some spell check feature, but when you have links in your SIG and other technical stuff in the email message it is a PITA. Much more efficient to notice the little red squiggly and pop the word into KWrite for magic to happen.

I keep seeing posts calling Gedit the Gnome equiv of KWrite. Well, it’s not. It has improved over the years, but it is still ugly. To compete with Kate you have to go commercial, or, you have to work around IDE needs in things like CodeLite and QtCreator just to write a blog post. The closest commercial product I have found is UltraEdit. There were a few years when I occasionally touched some nasty Windows stuff in the dark and distant past. (Thankfully I now only have to touch Windows once every 6 months or so to load results from my glucose meter for my doctor. Yes, enough sizable companies have left Windows so you never have to touch it again.)

During the dark period, when I had to touch that “other” platform, I wanted to find a commercial editor which ran on both (and ran the same on both) so I could load and go. That said, the Linux version was always the ugly red headed stepchild chained in the basement. The doc, help text, and even user forum answers would always be from a handful of Windows users. The real killer was the licensing. Ultimately they wanted $100/year for the rest of your life. Why pay that kind of money for something you don’t like as much as Kate and KWrite?

So, now that I’m going to have to use Gicky-Gnasty-Gnome for a while, I have to once again search for something which can begin to compare to both Kate and KWrite. At least until I can find a non-RPM based KDE distro which gets around the current set of NVidia problems.

Yes, I’ve looked at Jedit. I tried to use it for many months. Yes, there are a plethora of plug-ins. There are two things I’ve never been able to get over when it comes to that editor:

1) It’s a pig! To get many of the plug-ins to work you have to hack the JVM startup script and give it more memory which makes it a bigger pig. I can see why that “load in background at startup” option is so popular.

2) No matter what settings I tried the font rendering was simply atrocious.

I’ve tried almost every editor out there at some point. Kate and KWrite have simply been reliable tools which found places in my workflow. When I have to work in a situation where they either aren’t available or don’t work well, I become annoyed.

March 19, 2015

Dot-Com Flame Out 2.0

Anyone who was around for the last Dot-Com Flame Out (affectionately called DOT-BOMBS) remembers how that scam went. Valuations were mostly based on revenue growth (it was the new-new economy but old-old venture money). Groups of Dot-Coms started “trading services,” many in the form of advertising, to make the numbers exceed what the old economy investors wanted. Yes, a token few pointed out these companies weren’t actually making any money, but the hysteria over exponential revenue growth made idiots of rich and poor alike.

That was Dot-Com Flame Out 1.0, mostly. Eventually people began to realize these companies should have made money and should be able to pay their bills without more venture capital. And, well, your retirement account paid the price.

I believe I have discovered the vehicle by which Dot-Com Flame Out 2.0 will happen. In fact it is already in play. Even NPR has been running stories on this, but not quite nailing it. I wish I could have found a link to the one I listened to last night during the money report. The Wall Street Journal and others have been talking about eye-popping valuations and the various methods of arriving at them.

The valuation which is sticking in my mind is “eyeballs.” Yes, a good many are still buying the bull about the Internet being and advertising ecosystem so they are swallowing this eyeball scam hook-line-and-14-sinkers.

What scam you ask? Why only the newest and easiest non-cash way to raise money. Artificial eyeballs. You’ve all seen it, but you haven’t thought about it. Perhaps you thought “Oh, this is convenient, now I only have to remember one password,” but you didn’t actually think it through.

“What didn’t I think through” you ask?

Sign in using your FaceBook, Twitter, Google+ (okay, there are probably only two Google employees on Google+ so that really doesn’t count.) Many of the Web sites offering this “feature” are new and getting lots of buzz. The official word I got from one of them was they didn’t want to spend their precious startup money developing a login and validation. The dirty little secret is they can lay claim to millions of users the day the site went live even though not a single one of those users has ever been to their site.

Helloooo venture capitalist, have we got eyeballs to sell you.

Once the news of a multi-billion dollar valuation hits the press all of those “skilled” chart lemmings follow the stock over the cliff.