mancalablog

posts to days old
of topics
with text
post

Google

Paul (29 May 2004 2:12): I am pleased to report that in case you forget "manc.a.la", you can now get here by typing google.com/search?q=mancalablog&btnI. This also probably means Google will index links posted here, such as to MikeLog.
mike (30 May 2004 23:54): no, still no love from the google. they never bring me flowers.
Paul (14 Jun 2004 4:45): Darn it... now this page is no longer listed on Google either.
Come to think of it, wasn't there once a way to get Google to cough up the names of other sites that link to a given site? Is that still possible?
Grant Laughlin (15 Jun 2004 6:48): http://www.google.com/search?as_lq=manc.a.la&btnG=Search

But, nothing is listed...
The syntax is link:manc.a.la in the search bizox.
Paul (15 Jul 2004 18:42): Finally Google again lists both mancalablog and mikelog. A search for Mikelog Yale yields mancalablog as the second hit, with its link to Mikelog. However, a search for link:zoo.cs.yale.edu/~mjb63 does not yield any results. Am I doing something wrong?
m (15 Jul 2004 19:43): gah. I don't know. I've tried to find links to me but have never turned up anything. bzzzzht!
Paul (11 Feb 2005 18:44): Google Maps is pretty fun if you like maps. Unlike MapQuest and Yahoo, which appear to render maps dynamically, Google has prerendered all mapped areas at each zoom level, and just patches them together. This means a human could actually fix wandering labels etc...not that Google has gone that far yet...but we can expect better mapping to come, especially if competitors adopt Google's model. Now if only we could get international maps.....
(11 Feb 2005 21:32): Like for Japan. This whole 'streets don't have names' thing is kinda cool at first, but...
m (16 Feb 2005 10:55): when did google pick up blogger? I'd never noticed.
(16 Feb 2005 16:44): It was a long time ago - basically, when Google decided that it would begin to specifically index blogs.
m (16 Feb 2005 21:02): are the no-name posts all grant? Is he the only one who keeps fucking forgetting to put a name in with the posts?

link back to the original page with a button that says "post without name" maybe as a reminder?
zong (16 Feb 2005 22:48): As long as he is the only one...
Think of it as free compression for mancalablog =)
Paul doesn't much mind missing names...or...missing messages :-) (17 Feb 2005 16:53):
Grant (17 Feb 2005 23:31): I'm lazy and forgetful. Also, I'm scared that my coworkers who also read Mancalablog will notice my posts and then notice the timestamp and then realize that I'm posting during work and I'll be fired.
Still, was it really that hard to tell? What's with all the HATE? We're just trying to have a good time, NARC. We don't need your HATORADE!
m (24 Jun 2008 14:30): Anyone tried google maps lately from elinks?
No-map rendering! Just directions!
(25 Jun 2008 15:53): It's the same as always for me, because I don't have javascript or image-loading turned on.
m (26 Jun 2008 10:52): blink tag disabled, too.
G (26 Jun 2008 13:42): Yeah, unless you're running Netscape 3, or something. Kinda sad. :(

This guy's still angry, though.
G (26 Jun 2008 13:43): In other news, if you search for 'blink sucks,' all you get is stuff about blink 182, along with... so I guess 'blink' is an ability in World of Warcraft.
m (27 Jun 2008 10:45): Good idea, though, leaving implementation up to the browser with <HEINOUS> tags.
G (27 Jun 2008 16:01): Oh, god! That's horrible!
You guys probably can't see it, though, because your browser doesn't support <HEINOUS>.
m (30 Jun 2008 11:14): (don't worry, mancala sanitizes HEINOUS tags)
m (13 Dec 2008 0:32): What the fuck? Google screwed up its maps simple-interface, so now it no longer works with elinks. Damnit!
m (18 Dec 2008 2:19): yahoo does something right!
Paul (8 Jul 2009 5:19): ready for all apps to be web apps?
G (8 Jul 2009 13:21): No, not really.
m (9 Jul 2009 10:40): old idea
G (13 Feb 2010 14:34): I guess not all of you guys probably don't read my Google Reader Shared Items™, so here you go.
m (15 Feb 2010 11:06): Teabaggers Finally Google "Teabag"
Paul (24 Apr 2010 5:20): Upgraded to Lenny so I could get Chrome. It works as advertised---pretty fast even on my ancient machine.
R (25 Apr 2010 21:24): Chrome is pretty great. Key bindings are way better than Firefox, as is the speed.
m (28 Jul 2011 9:09): Grumpy? Grumpy!
those may look like random weird addresses, but they're run by google, so don't worry to hard. just log in to your google account to view them.
don't have one? click here!
Google (28 Jul 2011 15:10): They're run by google, so don't worry too hard.
m (1 Sep 2011 10:39): Forged Google certificate? SSL broken? Noooooo!
z (1 Sep 2011 13:48): So trust has been removed from DigiNotar, seems like the system is working.
G (13 Mar 2013 16:18): In a post to the company's official blog, SVP Technical Infrastructure Urs Hölzle said that Reader's numbers have declined over the past few years, and Google will thus retire the service on July 1, 2013.
m (13 Mar 2013 16:36): wait, really? that's unfortunate!
m (13 Mar 2013 16:37): aw, shit
G (13 Mar 2013 18:02): "While the product has a loyal following, over the years usage has declined."
I translate into:
We have no idea why people stopped using it right after we butchered it.
m (15 Mar 2013 16:28): what should I use when Reader is gone?
m (25 Mar 2013 9:15): rather sizable graveyard of dead google products, and a funny article (linking the graveyard) that draws a line from Google Reader to indispensable services that get taken over by the government (because: indispensable) and become public utilities. Interesting notion. Doesn't seem too plausible, but who knows? ICANN, IETF (ISOC), W3C - these are all private nonprofits whose services I could see being subsumed by some governmental body (maybe?)
m (25 Mar 2013 21:10): digg's doing a replacement for google reader? Interesting. But with weird value-added social components? Don't want that!
m (3 Apr 2013 21:11): Cory Doctorow gives a (tangential) shout to Newsblur as his likely RSS Reader replacement.
m (11 Apr 2013 9:10): Newsblur is impressive, but I'm not keen on seeing "premium feature" as often as I do. Too used to not paying for things. Might be a deal-breaker.
m (11 Apr 2013 10:16): Hm. Counter-argument: Don't be a Free User
m (11 Apr 2013 12:14): I like to think that Blink, the new in-house(?) rendering engine Google(?) is switching to for Chrome/Chromium is named after the unfortunate HTML tag.

Also: I'm clearly confused about Google's relationship to Chromium, to say nothing of Chrome vs Chromium.

(note: this whole post was just an excuse to do that with the link up there. fingers crossed that your browser listens to me and renders something so obviously not in your best interests)
G (16 Apr 2013 23:07): Yeah, doesn't work for me.
"I see what you did there," though.
m (14 Jun 2013 11:11): Our benevolent despots!
m (25 Jun 2013 7:21): need new RSS feed reader urgently!
m (25 Jun 2013 15:42): More rss readers: Feedly, The Old Reader, Bloglovin(??A?), and Vienna (on Ann's computer).
You guys try any of these / anything else?
m (25 Jun 2013 16:31): Feedly looks really slick, but their marketing guys seem a little too gung-ho
m (25 Jun 2013 16:56): so far I think I like The Old Reader the most, but it can be a little slow to load...
m (25 Jun 2013 17:11): CommaFeed?
m (26 Jun 2013 18:52): I don't know why, but the feed for mancala isn't updating in my TheOldReader -- maybe I made a malformed rss feed? Is there an RSS-validator somewhere?
m (26 Jun 2013 18:53): hm
R (28 Jun 2013 15:01): Yeah, this is getting dire. Feedly seems to be far and away the most popular, but they have no business model... which isn't great. Inoreader seemed OK at first, but checking the forums makes me think that the project is far too close to hobby status. Similar story with The Old Reader, it was slow and a little clunky a year ago, and it's not a commercial project. I keep pushing NewsBlur, but the "mark-things-read-after-two-weeks" thing is pretty irritating. Ugh. M, please give NewsBlur a shot real quick and see if you can convince me. I command you.
m (29 Jun 2013 8:52): NewsBlur was actually the first reader-alternative I loaded my feed-list into. But I'm not sold on it: the start page is noisy with all kinds of stuff I don't want, it loads slowly (as does everything I guess. google reader spoiled me on that), and what it does with my more-than-free-accounts-are-allowed number of feeds is confusing. It keeps the whole list of them but only updates some? What? I tried paring down to 64, but gave up. I guess you see this premium accounts thing as a feature, but I'm not so sure.

Feedly won't let you export an OPML of your feeds? Christ.

The Old Reader still refuses to update mancala's feed. I don't know why it hates me.
m (29 Jun 2013 8:56): The Old Reader is both slow and clunky. Agreed.

Also I remember Feedly asking to know too many things from my google account when I tried to register, so I never even made it past that step.

CommaFeed: definitely slow. But other than that I think I like.
R (1 Jul 2013 9:25): Well, today's the day. I think everyone has already done this, but definitely hit up https://www.google.com/takeout/ and back up your Reader data.

Then go to http://readerisdead.com and grab the very nice python scripts and run reader_archive. Looks like I have around 3.6 GB of feed data, so don't wait too long. OTOH, your internet connection is likely faster than mine. It took a couple of runs (it'll pick up where it left off!), but eventually succeeded, seems to like having almost 1 GB of real mem, fyi.

Looking for alternatives... I was checking out the Mr. Reader app blog, and he's got a some useful info 1 2 on the APIs that are out there. TL;DR: Newsblur API sux, and he's not currently supporting things that aren't reimplementations of the Google Reader API, which makes a lot of sense, bang-for-buck wise, since there are already a handful of sites, including Feedly, which implement that API.

Lastly, I tried Feedwrangler, and didn't really like the web interface much - too barebones. But Bazqux looks pretty good? There's a 30-day trial, so I guess it couldn't hurt. And if you get it going today, there's a star-import from a working Reader account. He's working on getting stars from the takeout export, but that's not done yet.

I guess that leaves Feedbin and FeedHQ if anyone has time to goof around further on this.
Sentence Clarity Police (1 Jul 2013 9:30): "TL;DR: Newsblur API sux, and he's not currently supporting things that aren't reimplementations of the Google Reader API, which makes a lot of sense, bang-for-buck wise, since there are already a handful of sites, including Feedly, which implement that API."

...
R (1 Jul 2013 16:38): Really liking the BazQux! It imports tags and stars! Super compact view options! Pretty much the exact same key commands you are used to for the Web! 30-day free trial, paid options from $9/year!

Also: It can SHOW COMMENTS IN LINE.

"What?" "YES." "Like, on Slog?" "YES." "AVOIDING THE SHITTY POP-UPS??!" "YES!" "YES!" "I TOLD YOU." "Yesss."

BazQux.com: I'm tired of dealing with looking for Reader alternatives.™
m (1 Jul 2013 17:47): Bah, gonna try BazQux. But I'm not into subscription-pricing!
m (1 Jul 2013 17:47): Also totally doing the reader_archive thing right now
R (1 Jul 2013 18:18): Well... just today Feedly implemented star import from Reader and OPML export? So, go that way if you like. As a back-end, it should have very similar client support to the rest of the Reader API clones, and it has a early start.

OTOH: Don't be a free user.

I predict Feedly will: Start adding advertisements / add "Premium Features" / sell your user data / start charging / go out of business. At least one. I'm not saying you need to pick BazQux — there's no sharing built-in, so it doesn't matter much — but I would have been happy to pay more than $9/year for Google Reader without sharing. And if they'd charged, it'd probably still be running.
R (1 Jul 2013 18:22): If it makes a difference, the Google permissions requested seem minimal, and they don't need or ask for CC information to start the trial.
m (1 Jul 2013 18:36): I understand the mechanics, and especially don't like that little free-to-use smorgasbord you just presented (and think you're totally right - at least one of those, if not more) - but that's only a certainty because of the hefty service they're providing caching/updating everyone's feeds, right? Should I just do that on my own locally with a piece of pay-once software? Is it that b/w intensive to poll a hundred feeds for modification time? I like the comments thing and love the community of slog commenters, but I never had that with google reader and can certainly live without if it means one fewer monthly fee.
R (1 Jul 2013 19:13): Ha, if you're willing to self host your back-end, there are definitely options out there: Fever ($30), tt-rss (free!), and newsblur (free... but jeezus-lord that looks ugly to stand up) all come to mind.
m (1 Jul 2013 19:32): Fever looks nice! The essential/supplemental dichotomy works well (maybe?) with how I think of my feeds. And I want a what's-hot sifting various entries to the top (which google did in its own way, I think just the feeds I checked most frequently).
m (2 Jul 2013 7:19): maybe we should get something running on grza?
m (23 Jul 2013 17:55): SAD KITTEN
G (24 Jul 2013 0:45): Cute cat, though.
m (29 Jul 2013 19:52): OLD READER NOOO
m (17 Dec 2014 15:56): Migrating from TheOldReader to Bazqux. Significantly happier with bazqux interface.
m (17 Dec 2014 16:28): Not thrilled about having recently-read posts vanish on me - only thing I can think to do is default to view-all instead of view-unread, but it doesn't look like there's even a keyboard shortcut for switching between the two (just the dropdown in the upper left) -- Reed, you able to help here?

Also: I want to default to list-view instead of expanded-view. Anywhere I can find settings like this? (will I get them after I start paying for a subscription?)
R (17 Dec 2014 18:09): Yeah, it does not appear that there is a quick and easy toggle for unread/all items. There is a mention of a UserVoice site to make feature requests. I've never done so, but the Dev responds amazingly quickly to help requests, he seems like a skilled and stand-up dude.

WRT list view - I think there isn't a way to set the default view, but changing it should stick - it's one of the options right below the search button.
R (17 Dec 2014 18:22): HEY M, CHECK IT
Thread Police (17 Dec 2014 18:24): smdh
m (18 Dec 2014 21:23): That filter/stream thing looks strong! And you're right - the button in the upper left was totally what I wanted.
R (4 Jan 2015 20:33): Hey, M! If you've moved to BazQux — do you have a tag that you are sharing? Right-click any tag, and there should be an enable button for making an RSS feed of the things you tag there.
m (6 Jan 2015 16:02): Neat! Just made 'starred items' into a public feed.
Paul (31 Jan 2015 6:23): Like a video game for linguistic nerds: Google Translate Community asks you to provide and rate translations. All the phrases it asks are absurdly short and provide no context---it explains this is just like the phrases people actually type into the Google Translate box, and it asks you to handle them according to what you expect to see Google provide as the resulting translation.

So I'm finding myself arbitrarily making up rules for how I think it should handle the resulting translation (this is for rating translations from Spanish to English):
* If the Spanish input can be interpreted either as a sentence fragment or as a complete question, then I expect it to translate as the complete question and add a ? mark. When adding a ? mark to the output, if the input was all lowercase, I accept either capitalizing or not capitalizing the first letter. (eg "que viste"->"what did you see?" or "What did you see?" but not "that you saw")
* If translating a word which English always capitalizes but Spanish doesn't, always capitalize. (eg "yo lo crei"->"I believed it" but not "i believed it")
* If the Spanish input is all lowercase, and the only reason to capitalize a word in English is for the first word of a sentence, yet the output hasn't added any punctuation to end the sentence, I disallow capitalizing the first word. (eg "hablame espanol"->"speak to me in Spanish" but not "Speak to me in Spanish")

I'll probably make up more rules as I go...
Paul (31 Jan 2015 7:22): ok, another capricious rule:
* If there's no additional clue, "su" should always be translated as "their" although it also accurately means "its", "his", "her", and "your" (formal). My extra justification for this is that "they" is taking over in English for all unknown-gender 3rd persons.
Paul (31 Jan 2015 7:36): * When the input is a 3rd person singular active verb with ambiguity of he/she, output the 3rd person singular verb still without a subject, although any of he/she/it are also accurate translations. eg "ya sabe"->"already knows" although it could be "he already knows", "she already knows".
Paul (31 Jan 2015 12:43): Ok so I got tired of Spanish so I switched to Latin. Wiktionary is good for Latin. Google is getting all the declensions and conjugations wrong. After much searching for each word, I'm putting "This user has disabled their account."->"Hic ūsuārius calendārium offēcit."