SoylentNews

About six months or so ago I decided to take a break from writing and do some reading, so I pulled an Asimov collection from the shelf. After half a dozen or so stories, I thought I'd read something that wasn't science fiction. Huckleberry Finn was on my mind, and since my copy was somehow lost I decided to just read it on the web; I remembered it being a really good book, though I hadn't read it in decades.

All the online versions sucked; archive.org, gutenberg.org, the .EDUs, it didn't matter. All had extraneous bullshit and looked thrown together carelessly.

So I thought I'd assemble one that didn't suck; one with borders, a book font, proper justification, all the illustrations by the original illustrator, and put there with thought rather than

.

I see why the rest looked hastily thrown together; it was a hell of a lot of work. I've been at it full time for a couple of weeks, and although I've posted it at my web site there are still a few bugs.

If you decide to read some Twain and see some bugs, please let me know.

The book isn't really about Huck. It was about Jim. It was an abolitionist book about the horrors of slavery, written before the Civil War.

Folks were sure different back then.

So I released the first update release for 1.2 series. It's a very small update, it fixes a bug that pissed me off and it adds support for something I wanted to get into 1.2.0.

The bug: There was a problem caused by dunce syndrome where the time for shutdown commands had extra or too few zeroes, caused by a tired coding problem.

The feature: Now the 'epoch status', 'epoch start', 'epoch restart', 'epoch stop', and 'epoch reload' commands can have multiple service names specified at once. That means instead of 'epoch restart aqu4bot;epoch restart aqu4bot_soy', now I can just do 'epoch restart aqu4bot aqu4bot_soy'. Saves a bit of typing if you have a big list of services you want to apply the same action to. The method I used to implement this for 'epoch status' is hideous and is ironic considering the goals of 1.2 Peroxide, but it works and it's not buggy or anything.

Those are literally the only changes in this release.
The 1.x.1 releases tend to be little bitty bugfixes and stuff.

Epoch Init System homepage
Download Epoch 1.2.1 (tar.gz)

For the last several years I've noticed Google's search results getting worse and worse as time went by. Ten years ago, typing the title of a work returned that work usually in the first spot. They now seem to completely ignore the "title" meta tags.

They've gradually reduced the number of things you could do to get to the document you're searching for. It used to be that if you searched for "no dog is a cat", all the search results would have that phrase in it, with documents having that as a title listed first. Now, searching for that phrase returns any page with any of those words. What damned good is that, if that exact phrase is what you're looking for? Apparently, Google now ignores quote marks.

I discovered this morning that that's not all it ignores now. Google has deteriorated to the point that the old Infoseek that Google took the search crown from was better.

It ignores not only quotes, but all punctuation and spaces. I searched for "Mars, Ho!" (including the quotes) and the first ten pages had results about people and things named "marsho". WHAT THE GOD DAMNED FUCK??? Why in the hell do the idiots think I put that comma, space, and exclamation point in for?

NASA has a "Mars, Ho!" page. Guess what? Google doesn't return it. At all. Google gives me millions of pages, none of which match my search criteria. I can almost see it not finding my insignifigant site, but a NASA page doesn't show up?

That's just pathetic.

What's even more pathetic is that its quality has deteriorated so much that Bing and Yahoo are now better at returning what you're actually searching for. Both NASA's "Mars, Ho!" page and mine are on the first page of both Bing and Yahoo's sites.

It was wrong of Firefox to just change peoples' default search, but I now see why they did it. They figured out before I did that Google jumped the shark quite a while ago and now is next to worthless.

The slashcode seems to make it's own mind up who my friends etc should be. I've not added ANYONE to any of those groups. Odd . . .

It's that time of year again. The time of year when everyone and their dog waxes nostalgic about all the shit nobody cares about from the year past, and stupidly predicts the next year in the grim knowlege that when the next New Year comes along nobody will remember that the dumbass predicted a bunch of foolish shit that turned out to be complete and utter balderdash. I might as well, too. Just like I did last year (yes, a lot of this was pasted from last year's final chapter).

Well, this one's starting out a little differently than previous ones. Is the whole damned internet down? It's Dec 17 right now and I was going to register the copyright for Mars, Ho! then work on his year's "Final Chapter". The copyright office is undergoing "emergency maintenance" and slashdot says "Slashdot is presently in offline mode. Only the front page and story pages linked from the front page are available in this mode. Please try again later" after Firefox warned me "It's a trap!"

As usual, first: the yearly index:

Journals:
the Paxil Diaries
2007
2008
2009
2010
2011
2012
2013

2014
A Pleasant Vacation
A Pretty Good Friday
Get off Wierd Al's Lawn!
The Coldest Night
A Yank Back to the Past
Scientist says white is black
I'm dreaming of a secular Christmas

Sci-Fi:
Nobots
Mars, Ho!
Moroned Off Vesta
Time flies like an error
Watch your language, young man!
Grommler

Last years' stupid predictions:
100% accuracy!

Someone will die. Not necessarily anybody I know...
SETI will find no sign of intelligent life. Not even on Earth.
The Pirate Party won't make inroads in the US. I hope I'm wrong about that one.
US politicians will continue to be wholly owned by the corporations.
I'll still be a nerd.
You'll still be a nerd.
Technophobic fashionista jocks will troll soylent.
Slashdot will be rife with dupes.
Many FPs will be poorly edited.
I'll retire and/or die.

Nailed it! I retired at the end of February.

I'll just keep the same list this year, except that the "retire or die" one (shudder) and replace it with "I'll publish Random Scribblings". That prediction may be a bit iffy.

Happy New Year! Ready for another trip around the sun?

It isn't supposed to be. I'll get to that later, but first, please download the Amazon e-book! It's only two bucks and I'd really like to see my name on a best seller list.

Speaking of names, the dufuses at Amazon insist on a first name. At least they left it lowercase.

As to its early release, I'm not sure what happened but I wanted to have perfect hard copies in my hand first, but I won't for another week. Not sure what happened, probably my own fault. If one of the covers is borked that's a $25 ISBN up in smoke. I can only hope.

As always, PDF and HTML is free. I'm only charging for the ebook because Amazon insists, and maybe it will get more exposure. I write these things to be read, after all, not for money. Good thing, too...

The free files and links to sales are here.

If you're thinking "hey, I already read that book, right here," well, no, you didn't. What's posted here is 2/3rds as long, and much of that was removed. If you can't afford two bucks, well, you can still read it for free.

Anyone that has used the login from the main page using the login dialog at the top-right will be silently surprised with the login being done using https but once completed it will switch to http. And thus dump your username association with your IP-address along with the session cookie for all three letter organizations to grab. Or your local hackers to abuse for impersonation.

There's a workaround for this. Simple go to any article from the main page. Click on the [Reply] link right underneath the article text. If you are not logged in ie missing session cookie. A login dialog will be added to your comment reply dialog. Just fill in username and password and click [Preview]. Viola.. logged in!

And don't forget ONE mistake on revealing your username association with IP, browser header etc.. It's there for a eternity or something bad happens to some harddisk racks in a certain location..

Happy SSL login to you! ;-)

I've been really busy the last few days. I sent for the (I hope) final version of the three physical formats of Mars, Ho!; when they get here I'll release the e-book to Amazon and the printed copies for bookstore sales. I'll need the URLs for the checkouts before I "officially" release it. It'll be a couple more weeks.

A while back I noticed "preview in web browser"in Open Office, so had a look. The paragraph indents were replaced by blank lines between paragraphs, but it was fully justified! So I saved it to my hard drive and opened it in Notepad. It seems there's an attribute to the

element I wasn't aware of;

gives you full justification. So I added it to the over 100 files for the two released books.

I'd used
      to separate paragraphs, which had worked perfectly in the files when they were left-aligned. A tiny bit ragged justified, but acceptable.

Then I thought "These pages need margins, dammit!" so did a little googling, and came up empty handed. So screw it, I went 1990s and used a one row three column table with everything except the navigation in the center column, with the left and right set to 5% of screen width each, and the center at 90%.

It looked really good; except now the paragraph indents were really ragged. So I went retro again and made a one pixel clear PNG and globally replaced
      with .

Also, if you have the Gentium Book Basic font installed it will display in that typeface (the printed book's typeface) instead of Times. I'm happy with it.

It's possible or maybe likely that there are some words that should be italics in the html versions of Nobots and The Paxil Diaries, since I just eyeballed it looking for italics and may have missed some. With Mars, Ho! I had Open Office generate its ugly HTML for each page so I could search for Voyage to Earth, too. It's not even novelette length yet, but is getting a bit long for a journal entry so I may start posting it bit by bit.

For the first time in nine years I got to see my youngest daughter on Christmas. Great Christmas present!

And logging on to Soylent I find ten mod points.

And the second to last pre-publication copies came Christmas eve eve. I finished going through it this morning, and the book itself is ready. What wasn't was the cover; I fixed it and ordered another copy, so Mars, Ho! should be online in a couple of weeks.

"Fourteen: The Final Chapter" will be posted New Years Eve. Its link to Mars, Ho! will take you to the "coming soon" page on my web site until I actually publish a few days later.

Merry Christmas, everyone. For you in Britain and Canada, happy Boxing Day (tomorrow). And to everyone, may you have a safe, well, pleasant, and happy 2015.

So, I've pushed out the final version of the Epoch Init System 1.2.0 "Peroxide". It's mostly a bugfix release and cleans up most of the nasty code in Epoch. Here's the tarball: http://universe2.us/epoch_1.2.0.tar.gz

I probably could have done more to it, but I wanted to just finish and get the release out so I could roll an update for my personal distro with the new version of Epoch in it. Nonetheless, this release should be pretty stable and is a recommended update. It is safe to update 1.0 and 1.1 releases to 1.2.0 without rebooting. Replace the 'epoch' binary with the new version and then run 'epoch reexec'. Check /var/log/system.log and it should tell you that you've been updated to 1.2.0. NOTE: if you do NOT run 'epoch reexec' after replacing the binary, your root filesystem will not be able to be remounted read-only on system shutdown and that could lead to data loss.
There is no good reason I can think of that you would not want to run 'epoch reexec'.

Here is a list of changes:

Changes since 1.1.1:

* Cleaned up a huge amount of code that was just fugly as hell. This is the big change.
* Removed unsigned long abuse caused by my (at the time) severe OCD.
* New service status output format. Looks cleaner.
* Extremely deprecated AlignStatusReports attribute completely removed. I doubt even one person will be affected by this.
* Add three new attributes: StartingStatusFormat, FinishedStatusFormat, and
StatusNames to manually specify an alternate service status output format.

* Specific bugfixes:
        * Don't set a config problem check to WARNING after we already found a FAILURE.
        * Fix overwriting service messages, caused by our old status format.
        * Fix inaccurate reporting of scheduled shutdown times, now report seconds too.