|
Tuesday, August 25. 2009
As former readers have known for some time, I've been wanting to write my own blog implementation. Finally, I've gotten a result which I'm happy with.
Ordinarily, I'd withhold releasing this information to the public, but since news of its existence has already leaked, I may as well let the cat out of the bag. It also helped that I made a not-quite-so-formal announcement at a recent SVFIG meeting.
So, if y'all are interested, please direct your browsers to http://www.falvotech.com/blog2/blog.fs where you'll find the world's first blog written entirely in Forth.
Some notes to remember though:
(1) It's still under development. So, don't expect certain features to work right away. For example, while it sports an RSS feed at http://www.falvotech.com/blog2/blog.fs/rss , no actual link to the feed exists yet. Why? I've just been lazy in the past. I'll probably be adding the feed link later this week or sometime next.
(2) It lacks support for comments. This is a long-term addition which requires significant planning on my part, because I want to balance the needs of my users against the needs of *me*, particularly in the face of all the spam I've been receiving.
(3) The current URL for the blog is its development URL, not its final URL. Therefore, the blog can break at any time, and for any duration, while I work to fix bugs or add new features. Those who subscribe to its RSS feed might find articles repeated often, as I alter its database schema or settings, like article IDs.
Overall, however, I'm quite pleased with the results so far, and will be investing about one hour or so per day towards its development. Note that this doesn't always mean new lines of code go in every day — I'm exercising formal verification while writing this code, so much of my time is spent with pencil and paper, writing and rewriting logic equations to assert the correctness of the code I do (eventually) write.
So, I hope you all enjoy the work-in-progress. Let me know via direct e-mail what you think!
Tuesday, May 26. 2009
Yeah, that's my dad.  Click on the photo for the full story.
Tuesday, May 19. 2009
My web application at http://www.rezurezu.com now supports the following features: - You are now provided permalink URLs to your resumes in HTML and PDF formats. Permalinks allow non-Rezu users to view the linked resume without the overhead of advertising.
- You can now download your resumes in plain-text, HTML, or PDF formats easily.
- I now have Project Wonderful advertising in place.
- Did I mention that rezurezu now supports PDF?
I have a few minor bugs to work out, thanks to those who reported them. But, overall, I'm rather pleased with my results.
Still coming is the ability to export content in Microsoft Office compatible .doc files.
I think I'll be taking a break from rezurezu to work on my new blog software. If worse comes to worse, I'll just can everything I've written before and stick with this s9y installation. But, I'm going to shoot for a somewhat different kind of blog -- one which, while it supports flat diary-type messages like this, is catered more towards proper, scholarly type articles instead. We'll see how things work out.
Monday, April 6. 2009
Job-hunting is an incredibly tiresome, boring, full-time, and unpaid job unto itself. Most of the work involved with seeking employment seems to come from sniffing out opportunities and from managing the plethora of resumés in your portfolio. While numerous web applications exist to automate the former, Rezurezu.com is the first web application to help automate the latter. Managed resumé hosting will help free up valuable time, time better spent networking with potential employers instead of with Office.
rezurezu.com is still in alpha right now — however, do not let this deter you from playing with the software. The application is stable enough now for general use. The following features need to be implemented before we take it to beta: - Ad-free resume hosting and download support for PDF, HTML, and plain-text formats
- The ability to re-arrange credentials after entry
.
Let me know what you think by e-mailing contact at rezurezu dot com. Thanks!
Sunday, March 29. 2009
It is with great regret, and yet a sign of relief, that I inform you that my blog is now down temporarily.
Well, yes and no. See, Serendipity is currently still installed. And as you can see, I can still write posts. So, no, the blog doesn't not exist. So what gives?
Well, in migrating my website from my previous hosting provider to my personal server, I've discovered that s9y cannot import blog entries from itself.
Huh?
That's right, kids! s9y does not provide a web-based API with which it can import articles, categories, media, et. al. from another s9y installation. Allegedly, you could perform an import through "generic RSS", which is what I've done to get what little content exists on the current blog instance. But, as you can see, Wordpress extensions produced numerous errors that broke the import, so I was forced to import without them. Thus, comments and categories, which I truly wanted, didn't come over.
Exporters into MovableType format are available, but each and every single one that I've found assumed MySQL database. Why is this so important? Because, s9y was configured to use PostgreSQL at my old hosting provider (since he didn't offer MySQL service at the time of its deployment), and PHP STILL, AFTER ALL THESE YEARS, LACKS A CONSISTENT, GENERIC DATABASE API. As a result, I couldn't even export into a portable format like MovableType. *sigh*
Then there is that small RSS issue that I noticed this morning, where I found a huge number of old articles were listed as new again. Why? These articles are all timestamped!
So, as you can see, rather than fight with a blog engine that is, while a complete joy to use as a user and as an author, nonetheless an administrative nightmare, I've decided to try my hand at writing my own blog software. So, I've closed down my blog. There will be no more articles posted here until the new blog goes up.
What will the new blog be written in? I haven't yet decided. I already have a functioning Django deployment on my server, so it seems logical that I re-use that infrastructure. And, yet, part of me wants to go off the deep end and write the new software in Forth, just for the hell of it, and just to see what is involved in writing it.
One thing is for certain: PHP is going away. I cannot and will not subject myself to the pain and suffering that is PHP. Any allegedly high-end dynamic web programming language that doesn't yet have a consistent database access API is, in my book, not worth its weight in slime-mold.
Tuesday, March 17. 2009
Can anyone please tell me why printing in Linux is so ###-damned hard? I've spent the better part of four hours so far trying to determine why my printer lists its status as "Printer not connected", when CUPS can autodetect the model and serial numbers. Care to tell me how this is possible if the device is not connected? Is there some fucking extra-sensory perception API I'm unaware of?
Look, even at its most primitive level, what you do is you take a stream of graphics commands (whether expressed as Postscript or not, I care none), you interpret them into a bitmapped image, and then you send that bitmapped image to the printer-specific device for rasterization. This is no more sophisticated than what the old-school AmigaOS operating system used (in fact, it was arguably more primitive, because AmigaOS forced you to print in "strips" if you lacked sufficient memory!). It's slow. It's bandwidth intensive. BUT IT WORKS!!! ALWAYS! WITHOUT USER INTERVENTION!!
In point of fact, even old-school, single-tasking GEM systems were apparently more sophisticated than what we find 25 years later with Linux. With GEM, they captured graphics output into a metafile, then stored the metafile into a directory, which served as the print spooler (appropriately named OUTPUT) input queue. Then, the OUTPUT desktop accessory would eventually get to your metafile, rasterize it, then dump it to the specific printer of your choice.
Now, I'm fully aware that CUPS, at least conceptually, works in a similar fashion. Send Postscript to a special socket, use Ghostscript to render it into some platform-independent bitmap representation, and then pass it to a device-specific daemon. The problem is, said daemon obviously is able to talk to the hardware (or else how is CUPS auto-detecting?); this narrows the problem down to the first two stages, which CUPS has explicit dominion over.
Look, if two 1984-grade operating systems that consumes less than 320KiB of memory each can print out documents generated by arbitrary programs, consistently, and with minimum fuss by the user, then so too can Linux! How is it possible that, after more than 10 years, printing under Linux still requires voodoo ceremonies and virgin sacrifices to Quetzalcoatl to get working? COME ON GUYS!!!!! THIS IS THE 21ST FREAKIN' CENTURY ALREADY!!!!
I am just SO sick of this already. 
Tuesday, March 17. 2009
Saturday, March 14. 2009
If you receive a phone call from 818-870-9185, either do not answer it or just hang up.
I had recently received an automated courtesy-type call from this number, informing me that my vehicle's warranty is about to expire, and it's the final warning before the case is forever closed. No mention was made of who was making the call, nor what vehicle ID, make, or model they had in mind when they called.
After some research, this number has a long history of scam involvement. The perpetrators involved also appear to be targeting cell phones as well, apparently knowing full well that their calls will consume minutes as a punishment for those who are smart enough to not fall for the scam.
Tuesday, March 10. 2009
According to this CNN article, signs now can detect their surrounding audiences and adapt their message. This brings to mind the retina-scanning signs and advertising devices used in the movie Minority Report. It's also just as frightening too — imagine a world where literally every surface is plastered with advertisements. Who wants to live in a world like this? Sure as hell not me.
Sunday, March 8. 2009
http://www.disclose.tv/action/viewvideo/19465/
A fascinating speech revealing that our contemporary problems have distant roots indeed.
Saturday, March 7. 2009
Due to the economy, my current hosting provider will need to close up shop. That means that falvotech.com (and thus, this blog as well) will need to find a new home. And that means, falvotech.com will be off-line temporarily while this happens.
When is the outage scheduled? Who knows? All I know is that this has to happen at least before the 31st of March unless other arrangements can be made.
To that end, I just might end up re-engineering the entirety of my website presence. I mean, if I'm going to go through all that work, then migration might prove more difficult than simply starting over from scratch. Again, no guarantees just yet — I'm still negotiating my options.
Just an FYI for y'all.
Friday, February 27. 2009
I came across the ultimate website today. It's an automated interface to Google. You literally can search anything on Google on behalf of someone else, and forward them the link. Be sure to read the real-time, helpful messages for the recipient too! And, it comes complete with a one-click interface to tinyurl.com too. How convenient is that?!
Thursday, February 26. 2009
An interesting concept, StupidFilter aims to programmatically classify allegedly stupid comments from public discussion fora. However, I don't think they define stupid quite like most people. From their FAQ: - Isn't filtering stupidity elitist?
- Yes. Yes, it is. That's sort of the whole point.
The answer to an otherwise succinct, valid, and honest question singularly defines what I'd consider to be a stupid response. Yet, their filter fails to detect it as such.
What would a non-stupid response to the aforementioned question look like? Here's my vote: - Isn't filtering stupidity elitist?
- While we feel that participation with a public discussion forum requires a certain minimum standard of both literacy and cogency, we ultimately cannot control how content providers use our product.
This wording leaves the reservation for the right to elitism with the stupidfilter.org ("While we feel that participation with a public discussion forum requires a certain minimum standard of both literacy and cogency"), while still ultimately placing the responsibility for its use on those intending on using it with their comment/article submission systems.
Here's another possibility: - Isn't filtering stupidity elitist?
- Like any tool, our product may be mis-applied by those who install it. However, we originally wanted to reduce the burden of the content provider editorial review process by identifying obviously grammatically incorrect text without human intervention. Such text appears regularly in link-spam comments, or comments intended to poison the corpus of a traditional Bayesian anti-spam filters. This frees the editorial staff to concentrate on more substantial feedback.
See? This response is also helpful. It provides some clarity into beneficial applications of the software, while conveniently eliding the author's true intentions (which nobody really needs to know about anyway).
I could go on, but I think you see where I'm headed with this. So far, taking samples from my own blog and feeding it into online demo, no actual stupid comment has successfully been identified. On the other hand, it has been spot-on in isolating both spam and people who are likely to be dyslexic (and, hence, either try to spell phonetically or otherwise have poor writing skills through no fault of their own) but who otherwise had good things to say. This leaves two avenues of applications for this software: - Isolate poorly written comments for editorial correction before posting,
- Isolate poorly written comments for immediate disposal, thus requiring commenters to maintain a minimum level of literacy and cogency, as hinted at above.
It's interesting that the former fails the "elitist" test, while the latter essentially defines elitism itself. As a tool, it can be used for either purpose.
Tuesday, February 24. 2009
I've got my towel. Do you?
Monday, February 23. 2009
Oh dear — the paper that started it all. Unix is a general-purpose, multi-user, interactive operating system for the larger Digital Equipment Corporation PDP-11 and the Interdata 8/32 computers. It offers a number of features seldom found even in larger operating systems, including i A hierarchical file system incorporating demountable volumes, ii Compatible file, device, and inter-process I/O, iii The ability to initiate asynchronous processes, iv System command language selectable on a per-user basis, v Over 100 subsystems including a dozen languages, vi High degree of portability. It's an interesting read. Allegedly, the original Unix environment only sported eleven (11) (1x10 1+1) system calls. Although I've not found a definitive list of these calls yet, I can believe it, considering the sheer simplicity of the original Unix environment.
|
