« first day (3481 days earlier)      last day (27 days later) » 

7:10 AM
Morning
 
 
1 hour later…
8:34 AM
@Vérace I'm still thinking about this, but even with the overlaps operator I'm stuck as it's possible to have row 1 overlap with 2, 2 overlap with 3 but 1 doesn't overlap with 3. So a gaps-and-islands approach is needed.
The other problem is you need to sort by both start time and end time to identify the new islands in case of short sessions that start later and finish earlier than a previous session.
 
8:59 AM
@Colin'tHart I've nailed it - I'm just writing up my answer now! Back in about 20 mins - i.e. 09:30 UTC!
 
9:16 AM
I have a query in my head too. Curious what yours looks like :-)
 
 
1 hour later…
10:23 AM
@Colin'tHart - answer posted. I'd be very interested in any comments which you (or any of our other esteemed "heapers") may have on a) the general approach and b) any efficiency recommendations which people may have.
3
Q: How to get non-overlapping distinct intervals from a PostgreSQL table?

ERJANUsing postgresql 9.6. The table has user sessions and I need distinct non overlapping sessions printed. CREATE TABLE SESSIONS( id serial NOT NULL PRIMARY KEY, ctn INT NOT NULL, day DATE NOT NULL, f_time TIME(0) NOT NULL, l_time TIME(0)...

 
11:15 AM
Some photos from my walk this morning... Feeding cygnets
Duck (mother) and duckling! There used to be 5! Mother must be young? Note to @AndriyM - _That_ is a duckling - note the differences! :-) And not ugly either! @mustaccio - also as "cute as a button!".
 
11:43 AM
I feel it's high time I did what I should have done from the beginning – JEAGL my tongue-in-cheek remark, as it's possible it might not have been understood as such. So here's full disclosure. I knew they weren't ducklings but (young) swans (and thank you to @Vérace for the words "cygnets" in any case). By calling them "ugly ducklings" I was referring to H. C. Andersen's fairy tale "The Ugly Duckling", where the titular character is actually a swan (a cygnet at the beginning).
Various other characters thought he was a duckling because he was born into a family of ducks. And as a duckling they thought he was ugly too, hence why they started calling him that.
And I called those lovely cygnets ugly duckling solely as a reference to the tale, not because I myself, too, consider them ugly.
There, I suppose my J is now EAGLed
 
@AndriyM - I suppose that I was being over-pedantic - of course you knew they were baby swans - and now you know the word "cygnet" (in Ukrainian молодий лебідь - what's the transliteration of the characters?). Originally from the Sanskrit -> Greek -> Latin -> Old French -> English. The modern French word for swan is "cygne" and "cygnet" is essentially "cygnette" - the "ette" suffix in French being a feminine dimunitive!
 
12:00 PM
@Vérace It's a little tricky, because I usually use y (as in yes) to transliterate й. But I also use use y to transliterate и, which is a sound similar to the English short i (as in pin). So that would make the transliteration of the phrase look like this: molodyy lebid'. The apostrophe in this case is to signify a palatalised d.
 
@AndriyM - so the "d" sounds kinda like a soft "g"/"j" - like the beginning of the French "je"? This is very common in Irish - the name "Bríd" for example is pronounced "Breege" - soft g!
 
I've seen -ий endings being transliterated as -yi (where i can also be used for the Ukrainian і), sometimes also -yj (j being uniformly used for й and y for и to avoid any confusion).
@Vérace Sort of, yes, it's as if you stop short of actually pronouncing the ʒ part of the phoneme
The b in lebid' is actually palatalised too, because it's followed by the i (similar in quality to the English long e sound; Ukrainian doesn't have the concept of length regarding vowels, though)
 
12:20 PM
@Vérace Perhaps a better example would be a word or phrase where /d/ and /j/ meet, like "deuce" or "would you". The palatalised d would be somewhere at the beginning of pronouncing the /j/ sound.
Or at the very end of /d/
 
12:55 PM
@AndriyM - more or less what I thought - similar to Irish! Almost like a "ssh" sound at the end of a word?
 
@Vérace Sorry, I'm not sure I understand what you are referring to there.
 
@AndriyM - the closest I can come to it in a commonly known international language is "je" in French, but kind of clipped at the end?
 
@Vérace The only relevance of the French "je" sound here that I can think of is that it indeed can help turn a preceding d into a palatalised d. But the sound itself is very different from the d, palatalised or not. Incidentally, Ukrainian has its own equivalent of the "je", which is ж (although the French sound is probably more similar to the palatalised version of ж). We usually transliterate ж as zh, by the way.
Or maybe I've totally misunderstood what you are talking about, sorry if that's the case
To be honest, I feel myself a little trapped in this conversation. I'm curious about various aspects of linguistics, a lot actually. But at the same time I'm by no means an expert. Add to that my unwillingness to leave you with a misconception mixed with my inability to explain adequately things that are out of my expertise :)
 
 
1 hour later…
2:18 PM
@AndriyM "I'm curious about various aspects of linguistics, a lot actually. But at the same time I'm by no means an expert." - we could be twins!!! Come forward and collect your prize (a big man-hug!) Maybe your Irish heritage which we discussed previously stems from the fact that you are in reality my long lost twin! :-) Seriously though, you've described (part of) me down to a "T"!
Google has a lot to answer for - I find myself starting to do some serious work and then before I know it, I'm reading pages about dead Ethiopian liturgical languages! I'm coming more and more to the belief that Nicholas Carr is correct - Google is making us stupid or keeping us in the Shallows.
Paul Graham has a good take on it here. Joel Spolsky (of this parish) also has words to say on the matter - separate offices for knowledge workers, only look at your email twice a day. All good advice, I just wish that I could follow it! :-)
 
I wish I could too! :)
@Vérace Yes, my hidden Irish heritage seems to manifest again. Although I should also add that last time I was thought to be a lost twin brother of your brother-in-law. This time it looks like I might be your lost twin.
Maybe it's how being a lost twin works. Maybe you can be a lost twin to many different people
 
@AndriyM If my you, my brother-in-law, you and I were all triplets who have found each other at last, that would fit the scenario. I don't look forward to informing my sister though! :-)
 
2:33 PM
That conversation would likely be more awkward than my attempts to explain linguistic terms.
 
 
1 hour later…
3:51 PM
@Colin'tHart - does my approach to the question match yours - I'd be interested to know what you were envisaging!
 
 
8 hours later…
11:49 PM
Reminds me a bit of the recend moderator crisis on one of the popular web sites...
"recent", of course. Rosé is taking its toll...
 

« first day (3481 days earlier)      last day (27 days later) »