Still up, reading conference proceedings. Not finding much- apparently that's good. Better would be not doing this at almost 2 am. Ah, the dissertation.
Clever phrase: cute elaboration
This is the title pattern. I'm running into scores of article using this same tired format, a tiny subset of which are of any interest to me. I'm also finding in the Zotero-Sqlite-ooBase link that some of the oddities just become part of the workflow- like having to shut down Base before making updates to Zotero, since it tends to freeze up when trying to access Sqlite. My bad, Sqlite was never intended to be a shared application.
Someday I'll make a list of what I've learned about this setup- which isn't quite as bad once you get used to it, or at least nowhere near as bad as the alternatives. But, I realized I was double-counting the stuff in the Zotero "trash" folder, which threw off my counts quite a bit.
At the end of the day, it's still clear that in the areas of transportation and Personal Information Management / Task Management that I care about, only a handful have knocked out more than a couple of research papers, and I don't have any sign that anybody's doing anything interesting with data structures than I knew about initially.
One side of the coin is that I just wasted a month confirming what I already knew. The other side is that I'm better able to back up what I already know. This ain't a Harry Potter coin, so it's finally off to bed to dig through more papers tomorrow.
Monday, July 27, 2009
Friday, July 24, 2009
Tell me a story
Getting a coffee at Uncommon Ground last night, mostly to justify occupying a table and tapping free wifi for a couple of hours. The guy behind the counter had a nametag on that read, "tell me a story!". I asked- he mentioned that all his coworkers talk about is going out, drinking, and various other 20-something mayhem that goes for social life in Albany. (I think I was young once, too, but that was too long ago to remember for certain.)
So, he asked, do I have any stories involving "real jobs"?
Offhand, I actually don't. Nothing the general public needs to know about. But something did pop into mind: the day I brought my daughter to visit my old job. So, long story short, I brought her in, stopped in to talk to my old boss to say "hi", who promptly picked her up, only for her to poop her pants, and it leaked all over on him. Then she did it again in the elevator as we headed home.
Actually it didn't quite happen that way, and I don't know why I lied about it. Well, maybe I do.
It was Jan 2, 2001, I'd recently left my job in the World Trade Center to work Uptown. On request, I brought my 4-month-old daughter to visit my old floor, 72, on WTC1, the one more famously called the "North Tower". (I left out that part- you can't talk about the WTC without everyone remembering 9/11. FWIW, I think of the building on 9/10 and before, the last day I set foot in there.)
We saw my old boss, who waved to her, and told me to cherish every day, because babies grow up way too fast. Then of course, my daughter did load her Huggies on the Chief Engineer's sofa. I saw my old cubicle, not 20 feet away from there, now converted to a storage area for boxes, broken chairs, and empty tape dispensers.
But at my former co-worker's cubicle was my old computer, my other old computer, and the other of my other old computers- 3 of the 5 I had in there, the other two were moved to a work facility across the river. My array of post-it-notes was transplanted to my co-worker's desk in pretty much the same locations as in my office.
My old router was there- unused. Back in the day, the network administrator gave up on whatever I was doing, handed me a router, and just told me to keep my stuff from touching "his" network- the array of junk machines I converted to application servers that saved dozens of hours of batch processing a month that used to virtually shut down the network. Since I'd left, they'd reintroduced my servers back onto the network, though they'd shelved my internal web server project. Had a quick talk about the database server we'd just started- the one I almost stayed on for before moving to a lesser job for better family health insurance.
Then of course, I finally left, since babies and debugging database server code don't mix well. On the way down the elevator, my daughter once again filled her diaper, and was changed on a bench on the side of the main concourse of the WTC mall. Then it was down the escalators, back onto the PATH train, back to the parking garage in Newport Mall in JC, then back home.
The next time I spoke to my old boss, it was 9/12, and I called his house to make sure he was still alive. It was a few days after my daughter's first birthday. He thanked me for my concern, and told me to just be glad I'd changed jobs- though they didn't like to see me leave, he was glad I was nowhere near there the day before. I could have, because I used to change trains in Hoboken, and it was a coin toss whether I went to the WTC or to 33rd street- it depended on which train had empty seats when I got to the platform. And on that day, the WTC train was full, and I didn't feel like standing. Apparently, it arrived at the terminal shortly before the first plane hit.
I don't know why this story came to mind then, and why I'm staying up late to write it. There are a lot of other stories to tell, but my time working in the WTC which was the high point of my career, quite literally, in my 20's in the hope, optimism, and even arrogance of the pre-bubble-burst dot-com era, going to Chicago once or twice a month, and trips to Japan, Spain, France and Italy on my own dime were possible. Back when writing a little SQL was brain surgery and hacking out HTML by hand was like rocket science, the sky was the limit as I was plotting my own venture capital-backed dot-com. But all that was so 9/10.
That's the real story. How I ended up back in Albany is a far longer one.
So, he asked, do I have any stories involving "real jobs"?
Offhand, I actually don't. Nothing the general public needs to know about. But something did pop into mind: the day I brought my daughter to visit my old job. So, long story short, I brought her in, stopped in to talk to my old boss to say "hi", who promptly picked her up, only for her to poop her pants, and it leaked all over on him. Then she did it again in the elevator as we headed home.
Actually it didn't quite happen that way, and I don't know why I lied about it. Well, maybe I do.
It was Jan 2, 2001, I'd recently left my job in the World Trade Center to work Uptown. On request, I brought my 4-month-old daughter to visit my old floor, 72, on WTC1, the one more famously called the "North Tower". (I left out that part- you can't talk about the WTC without everyone remembering 9/11. FWIW, I think of the building on 9/10 and before, the last day I set foot in there.)
We saw my old boss, who waved to her, and told me to cherish every day, because babies grow up way too fast. Then of course, my daughter did load her Huggies on the Chief Engineer's sofa. I saw my old cubicle, not 20 feet away from there, now converted to a storage area for boxes, broken chairs, and empty tape dispensers.
But at my former co-worker's cubicle was my old computer, my other old computer, and the other of my other old computers- 3 of the 5 I had in there, the other two were moved to a work facility across the river. My array of post-it-notes was transplanted to my co-worker's desk in pretty much the same locations as in my office.
My old router was there- unused. Back in the day, the network administrator gave up on whatever I was doing, handed me a router, and just told me to keep my stuff from touching "his" network- the array of junk machines I converted to application servers that saved dozens of hours of batch processing a month that used to virtually shut down the network. Since I'd left, they'd reintroduced my servers back onto the network, though they'd shelved my internal web server project. Had a quick talk about the database server we'd just started- the one I almost stayed on for before moving to a lesser job for better family health insurance.
Then of course, I finally left, since babies and debugging database server code don't mix well. On the way down the elevator, my daughter once again filled her diaper, and was changed on a bench on the side of the main concourse of the WTC mall. Then it was down the escalators, back onto the PATH train, back to the parking garage in Newport Mall in JC, then back home.
The next time I spoke to my old boss, it was 9/12, and I called his house to make sure he was still alive. It was a few days after my daughter's first birthday. He thanked me for my concern, and told me to just be glad I'd changed jobs- though they didn't like to see me leave, he was glad I was nowhere near there the day before. I could have, because I used to change trains in Hoboken, and it was a coin toss whether I went to the WTC or to 33rd street- it depended on which train had empty seats when I got to the platform. And on that day, the WTC train was full, and I didn't feel like standing. Apparently, it arrived at the terminal shortly before the first plane hit.
I don't know why this story came to mind then, and why I'm staying up late to write it. There are a lot of other stories to tell, but my time working in the WTC which was the high point of my career, quite literally, in my 20's in the hope, optimism, and even arrogance of the pre-bubble-burst dot-com era, going to Chicago once or twice a month, and trips to Japan, Spain, France and Italy on my own dime were possible. Back when writing a little SQL was brain surgery and hacking out HTML by hand was like rocket science, the sky was the limit as I was plotting my own venture capital-backed dot-com. But all that was so 9/10.
That's the real story. How I ended up back in Albany is a far longer one.
Saturday, July 18, 2009
Bad software. Very Bad.
I'm discovering that making a query from existing queries in Open Office Base doesn't work well- it's buggy. It's also not picking up changes from an SQLite connection- so something clearly is wrong here- Zotero saves in SQLite, but there have been a couple of database restructurings since then, and there may be some incompatability there. Still, if they're using the standards, table/field changes shouldn't matter.
I'm wondering why Blogger doesn't make it very easy to create links to prior blogs- or anything else for that matter. Granted, I can open another tab in Firefox, search for the link I want, copy the URL, and then paste it into the Make Link window based on the text I've highlighted as the target for the link. Yes, easy for somebody who started in IT in the early 90's. But annoying?
At my summer gig, I discovered that the text output from SQL Server Bulk Copy outputs a different text format from the XML output from SQL Server Field Format files as rendered through XSL. After a few hours of trying to match code pages, hex editor scan the pages for extraneous bit flags, etc, I finally realized that I could just use MSDOS "TYPE" to stream the output into a third file from the first and second file, then discard the first and second files. I hadn't had to do a clumsy hack like that since, say, 1997. Nobody in the past 12 years has had to attach field names to database table output? Seriously?
And all that's because SQL Server's Excel output seems to crash above about 3000 records. Like a real database has less than 3K records- I used to work on systems that would kick out 250K records a month or more, and the last big one I worked on had tables of 60M records. Not that you'd dump that into Excel, but I'm just sayin'....
These are the days when I agree with an old poster: if civil engineers built cities the way software developers wrote software, the first woodpecker to come along would level civilization. {sigh}.
I'm wondering why Blogger doesn't make it very easy to create links to prior blogs- or anything else for that matter. Granted, I can open another tab in Firefox, search for the link I want, copy the URL, and then paste it into the Make Link window based on the text I've highlighted as the target for the link. Yes, easy for somebody who started in IT in the early 90's. But annoying?
At my summer gig, I discovered that the text output from SQL Server Bulk Copy outputs a different text format from the XML output from SQL Server Field Format files as rendered through XSL. After a few hours of trying to match code pages, hex editor scan the pages for extraneous bit flags, etc, I finally realized that I could just use MSDOS "TYPE" to stream the output into a third file from the first and second file, then discard the first and second files. I hadn't had to do a clumsy hack like that since, say, 1997. Nobody in the past 12 years has had to attach field names to database table output? Seriously?
And all that's because SQL Server's Excel output seems to crash above about 3000 records. Like a real database has less than 3K records- I used to work on systems that would kick out 250K records a month or more, and the last big one I worked on had tables of 60M records. Not that you'd dump that into Excel, but I'm just sayin'....
These are the days when I agree with an old poster: if civil engineers built cities the way software developers wrote software, the first woodpecker to come along would level civilization. {sigh}.
Friday, July 10, 2009
Liking OpenOffice Base and stuffs
So, taking a break from taking a break and finally back to firing queries into Zotero to help clean up my citation list. I'm still wary of using SQLite to update Zotero- just reading data is fine for now. I prefer to use the native application for modifying data, since you never know what data integrity rules are buried in the client app.(Momentary shudder while remembering that decade of fixing other people's database apps.)
But OOo Base looks a lot like MS Access, so if you just happen to be familiar with the latter, you can use the former. I hadn't bothered to use Base before, but once in it, it seems easy enough. But these open tools allow a lot of hacking you can't do with the commercial stuff, like tap into the data store with downloaded drivers and free tools off the Internet. I don't know if this stuff has been done before, but I haven't found anything about it via Google. And I didn't have to mess around with RDF or XML / XSLT stuff- those would require discrete data exports, and I want to work with live data.
First thoughts about True/Slant
With my in-laws on the plane back to Tokyo, I'm looking at the remnants of a literature review for next week, a summer gig doing database work, and my son with a tummy ache- which I really hope isn't the latest round of some GI bug I had last week that knocked me and my daughter out for a day. But I'm anchored at home today, with the kids relaxing a bit in front of Charlie and Lola while I get back into writing.
I've noticed an interesting Augmented Reality post in one of my new online news reads, True/Slant. They run some interesting bits, but the number of columns is rapidly growing and it seems harder to stay on top of the articles. Though they seem to be on to something here- the surprising range of content- but the navigation could use improvement. There's nothing wrong with it- quite conventional in fact- but I'd like to see something thoroughly unconventional about the way content is organized.
Some thoughts:
I've noticed an interesting Augmented Reality post in one of my new online news reads, True/Slant. They run some interesting bits, but the number of columns is rapidly growing and it seems harder to stay on top of the articles. Though they seem to be on to something here- the surprising range of content- but the navigation could use improvement. There's nothing wrong with it- quite conventional in fact- but I'd like to see something thoroughly unconventional about the way content is organized.
Some thoughts:
- I think they could use a few different interfaces- like a Social Network view of some sort, between authors, and between topics.
- They rely on tags and RSS a lot. It's Web2.0, with interesting visuals, but needs a little more punch.
- More delivery channels- like iPhone, or PDA, or Kindle- not just plain ol' browser. But something offline would be nice- in case you want to read without Internet access. Google Gears?
- This could use other data exports- not just RSS. Like whatever the heck xml Google uses for calendar, maps, etc.
- Tighter integration with TinyURL (or their own self-branded version of it) and Twitter.
- The format is author / article / link / discussion board. I didn't realize how tiring that is until now. Somehow it just doesn't quite fit the site, but I'm not sure what would work better. It's just what everybody else does.
Wednesday, July 01, 2009
24 hour bug
I'm sick with some kind of (hopefully) 24-hour GI bug. My inlaws are in town, and let's just say that a GI bug and loads of kimchee don't mix well. Funny how these things drive you to childhood treatments- mine included toast, gelatin, and flat ginger ale.
On a brighter note- I'm experimenting with Zotero, trying to get at the data store with an ODBC driver for SQLite. Unless I'm doing something wrong, there are loads of empty tables, or tables full of primary/foreign keys- nothing terribly useful yet, but on the right track. I may have finished cleaning up my reference data, or at least the stuff I care about. What I want is SQL access to the raw data- with a couple hundred articles, and several hundred authors, it would save an enormous amount of time to find the publication history of contributors.
Off to more Jello and Meekrat Manor. I can't tell the meekrats apart, and I really don't care to.
On a brighter note- I'm experimenting with Zotero, trying to get at the data store with an ODBC driver for SQLite. Unless I'm doing something wrong, there are loads of empty tables, or tables full of primary/foreign keys- nothing terribly useful yet, but on the right track. I may have finished cleaning up my reference data, or at least the stuff I care about. What I want is SQL access to the raw data- with a couple hundred articles, and several hundred authors, it would save an enormous amount of time to find the publication history of contributors.
Off to more Jello and Meekrat Manor. I can't tell the meekrats apart, and I really don't care to.
Subscribe to:
Posts (Atom)