« first day (2345 days earlier)      last day (2681 days later) » 

6:00 AM
Not too sure with openSUSE...
 
(I don't have any Skylake hardware to test it on)
 
is 3 pane
so, essentially, its "I want full articles or partial articles, for the feed or category I am on to the left"
exactly as dog intended it ttrss or the late google reader did
 
Bob
@JourneymanGeek oh. is there even much point in doing an app, then? o.O
what would an app have that the web ui is missing?
 
@Bob if the app actually would let you download stuff for offline use...
the android app does
 
Bob
6:01 AM
@JourneymanGeek ah. hmmmm.
 
and er. sometimes you just don't want everything on a webpage
and you could have a standalone reader that has an option of talking to ttrss
 
Bob
that sounds like something... that doesn't really interest me personally
I mean the offline use thing
:P
 
and well.
seriously. 3 panes suck
everyone does 3 panes.
@Bob the baseline rss feed reader would save things locally anyway
 
Bob
@JourneymanGeek that would be the thunderbird way, I believe
 
@Bob its inefficient
 
Bob
6:03 AM
@JourneymanGeek when you say save locally... iirc rss was mostly just summaries, but you'd want the full page to be fetched, I suppose?
 
Bob
HMMMMMMM.
 
add that and searchability
and essentially you have a lovely external brain on a stick
 
Bob
@JourneymanGeek ok, now that is a tall order :P
(search is surprisingly hard)
 
Bob
6:04 AM
@JourneymanGeek actually, I think the whole concept of offline mode would be more interesting as a "download to server" rather than "download to computer"
 
Its not a minimal requirement
 
Bob
i.e. it would be better handled on the ttrss server
 
@Bob but ttrss downloads things to itself anyway
 
Bob
oh, it does? o.o
@JourneymanGeek so if the site deletes a page... ttrss keeps it?
that sounds useful
 
er
I don't think it keeps the images
and periodically purges
 
Bob
6:06 AM
aww
:\
I suppose I could use Lucene.Net for searching
never used it before :P
 
oh, you could disable it
but the author will mock you if it breaks anything
this is a common theme with ttrss
 
Bob
...last released 2012
 
see what SE uses ;p
 
Bob
@JourneymanGeek iirc Elasticsearch
which is vastly overkill
 
ah
also somewhat finicky
 
Bob
6:08 AM
> Elasticsearch is a search engine based on Lucene.
WELP.
213
Q: A new search engine for Stack Exchange

Nick CraverAfter the performance problems we have run into with Lucene.NET we've decided to make a change, we're moving the network on to elasticsearch. Here's where to get started: http://stackoverflow.com/search What works: All search operators should be in Many changes below from the old search behav...

@JourneymanGeek ^
 
Bob
> After the performance problems we have run into with Lucene.NET we've decided to make a change, we're moving the network on to elasticsearch.
welllllllllllll
 
guess not
 
Bob
that's literally the same option I mentioned. twice.
 
lol
this does feel like feature bloat tho. ;p
way beyond minimal/shippable.
 
Bob
6:10 AM
@JourneymanGeek bleh. I don't really feel like doing a proper MVP
and eh
I have half a dozen other projects I should technically do first
 
;p
not that interesting?
 
Bob
@JourneymanGeek remind me in a month? :P
 
(RSS is kind of a deep geek tool these days)
 
Bob
@JourneymanGeek actually, I really want to set up RSS (again). But I should also finish up that thing for @allquixotic, and finish up the soundboard thing I started last week.
 
Bob
6:11 AM
@JourneymanGeek There's lots of sites I currently keep track of by manually visiting them. RSS would be nice.
 
... actually
 
Bob
I used it back in... uhh. 2010? with Thunderbird but it became unmanageable pretty quickly
also I kinda need phone support, so I suppose I will be going with ttrss :\
 
Its dead but
ttrss's android client isn't free, cause the dev hates coding it ;p
 
Bob
lol
 
^ sputnik's the closest thing to a 'perfect' desktop client I've used so far
 
Bob
6:12 AM
sounds like my kind of dev :D
 
not sure about search, but it dosen't have ttrss support
 
Bob
@JourneymanGeek y'know. It's written in JS + nwjs.
 
@Bob god, the forum RSS feeds are 'stock' and filled with the smell of burning n00bs
 
Bob
That's literally my current job.
 
@Bob ;p
 
Bob
6:13 AM
@JourneymanGeek Is there any particular problem with it being dead?
 
I don't think JS belongs on the desktop/server
 
Bob
like, something you want to add to it?
cauuuuuuse.
 
@Bob no, I'd be using it if not for...
ttrss support, maybe themes.
 
Bob
@JourneymanGeek well. I'm terrible at themes. But it's just CSS here.
dunno how hard ttrss support would be to hack in
it's on nwjs 12.0, which is a bit of a problem
cause 13.0 was a major breaking release
@JourneymanGeek there's also 74 forks of it, but they add nothing :P
ewwwww
NSIS installer
let's see what the code looks like
urk. angularjs. oooooooookay. I know JS. I do not know Angular.
 
lol
I actually was meaning
 
Bob
6:17 AM
hm. wonder if I can be bothered figuring it out :P
can't be that different from Ext
 
"If you wanted an RSS feed reader and didn't want to set up ttrss, this is nice"
;p
 
Bob
hm. it's well-written
@JourneymanGeek in fact, this is probably better than ttrss for me, since ican definitely hack JS much better than PHP
but it's not synced, so it'll be a bit of a pain between desktop/laptop
 
which is why I like/use ttrss
 
Bob
ugh
ok
tonight
finish @allquixotic's thing.
 
Bob
6:19 AM
then, RSS time :P
also why is chat giving me 10s timeouts :(
 
One of these days I need to work out if I should feel guilty for throwing you these odd little quests, or good that you find projects that are fun ;p
 
Bob
@JourneymanGeek the latter :P
I'll probably start with standard ttrss w/ web ui
and think about doing a desktop thing later
would be a good excuse to learn UWP actually
 
oh
and if you set up your own, and use the dark theme
I have some CSS for stopping it from being monochrome.
let me know if you need it (or you need it for your account on mine) so you can laugh at my terrible hack-job
 
Bob
@JourneymanGeek I'm usually fine with light themes :P
@JourneymanGeek heh, if it's actually horrible I could clean it up some. but it's probably fine
 
lol
its not horrible. maybe
 
Bob
6:23 AM
(horrible CSS hacks usually include overly-broad or overly-restrictive rules, or selectors depending on element positioning when there's alternatives)
 
Bob
i.e. use classnames (and possibly static IDs) over things like "4th down"
 
(you stick this in somewhere to override the theme)
I don't even remember if I copied it or beat it into working
 
Bob
@JourneymanGeek ehhhh looks alright, I assume the specific selectors are to override the original
 
Bob
6:25 AM
cause they're a tad overly-specific but sometimes that's necessary because more-specific overrides less-specific
 
pretty much
 
Bob
(that's why it's better to do "least specific necessary" so later overrides are easier... if you make your original too specific then !important gets brought in)
as an aside, !important is the CSS equivalent to the ol' regex quote :P
 
I am not very fond of the Intel Inside program.
 
Bob
"I had 99 CSS specificity problems, so I used !important. Now I have 99 !important problems."
 
6:28 AM
Among other things, Intel requires OEMs in the program to use slogans like "Intel Inside. Amazing Experiences Outside." (and similar) in their marketing.
 
Yeah, you hate intel, we get it
 
I just want to see AMD get a fair shot at the market.
nvm
 
@Bob actually, I suspect the reason I want a desktop client is tradition, and occationally being in places with suspect internet
but that's gotten so much better
 
My apologies.
AARGH! Why do I keep making mistakes like this?
!!/ultraheaddesk
 
6:32 AM
heh
@Bob the more I think about it, the more it feels like a non-problem XD
 
I guess my soft skills are just too underdeveloped.
 
Bob
@JourneymanGeek I think the hard part with offline is... well, the whole thing :P
@JourneymanGeek how's ttrss's search, btw?
 
good enough
 
Bob
@JourneymanGeek ...'enough' implies it's lacking somehow? :P
 
I don't use it as much as I did google reader's search
Its kinda in a filing cabinet in a toilet...
 
Bob
6:35 AM
@JourneymanGeek well, I could try slapping Lucene on it and see what happens :P
 
lol
meh. To actually have good search, the app needs to be fundamentally different ;p
 
Bob
@JourneymanGeek how so?
@JourneymanGeek tbh I've never done any proper indexed search before so this'll be a good experience :P
even if it is just leaning on existing libraries
 
@Bob ttrss is designed more as a way to consume feeds
and well, fox claims the backend isn't going to be very happy with too many feeds
and well, a few google reader clones have gotten backed up and explodinated.
google reader essentially had unlimited resources and was designed for search
 
Bob
@JourneymanGeek bleh
I should write my own uber-inefficient one in ASP.NET and chuck it on a $100 dedi :P
should be enough horsepower behind a 4-core Ivy Xeon, no?
 
lol
Actually, I wonder if this is one of those things a non traditional database would handle better
and lol
that's reminding me of a thing I actually need.
I'd like something where I can give it a URL, and an element, and generate a rss feed when it changes, for those webcomics with no RSS feeds ;p
lol. I wonder...
sounds like the sorta thing I could throw up if I tried programming. SEEMS simple.
 
6:52 AM
0
A: If you're gonna talk Politics, you must respect those who disagree

bwDracoThis extends to abuse of moderation tools like flags. Seriously, don't flag stuff simply because you disagree with the other person's opinion. A while back, I responded to a series of chat flags raised on The Bridge, Arqade's main chat room and perhaps the most active room on Stack Exchange. I...

Not my usual writing style.
 
Bob
@JourneymanGeek in all seriousness? depends how the search is implemented
 
@Bob and I suppose the content.
 
Bob
@JourneymanGeek not really.
The search engine (Lucene, etc.) will have their own ways to store indexes
 
Bob
@JourneymanGeek Not simple.
It's effectively polling.
Oh, I've done it to watch for stock level changes.
 
6:56 AM
@Bob well, I don't need it in realtime
 
Bob
It's theoretically very simple.
Just... polling is ugly.
 
technically RSS polls too
you're checking for a new updated feed every say 15 minutes
 
@Bob Yup.
Problem is, push needs a lot of extra work on the server side...
 
Bob
@bwDraco tbh it's not much work
it's just that most servers won't do it unless explicitly needed
 
nothing else
its a basic XML file
 
Bob
6:59 AM
@JourneymanGeek yea, but it expects it
 
that's essentially the beauty of RSS tho ;p
the client works out what to do with it
 
Bob
polling a service that doesn't expect it can lead to throttling and blocking
possibly.
depends on frequency of course
every 5 mins is cutting it close
 
Push requires the server to know what clients it needs to serve. Pull requires no registration.
 
Bob
15 mins is probably fine
 
Bob
7:00 AM
half day is fine
 
I'm thinking more once or twice a day
 
Bob
@bwDraco kiiiiiiiiiiinda
 
and I'd probably hook it into cron
 
For RSS, I'd want to say hourly.
 
Bob
there's also long polling
 
7:01 AM
Email clients have always polled servers...
I have Thunderbird set to hit Yahoo Mail every 15 minutes, and that's on three accounts.
 
Bob
@bwDraco Uhh. No.
POP3 was polling-only.
POP3 also sucked in every way except maybe simplicity.
 
Hmm...
 
Bob
IMAP does not poll by default.
 
I have not really used IMAP anywhere.
 
7:03 AM
Thunderbird sets up Yahoo Mail with POP3 by default for whatever reason.
 
Bob
@bwDraco ewwwwwwwwwwwwww
POP is absolutely shit.
 
IMAP is awesome less sucky
 
(except on my tablet)
 
Bob
No folder support. No sync support (it's download-only).
In email technology, IDLE is an IMAP feature described in RFC 2177 that allows a client to indicate to the server that it is ready to accept real-time notifications. == Significance == The IDLE feature allows IMAP email users to immediately receive any mailbox changes without having to undertake any action such as clicking on a refresh button, or having the email client automatically and repeatedly ask the server for new messages. == Usage == IMAP4 servers that support IDLE will include the programming string "IDLE" in the result of their CAPABILITY command. This allows email users to r...
 
It does make it easier to maintain permanent, long-term email archives...
 
Bob
7:04 AM
FYI
IDLE is supported by pretty much every IMAP client and server in existence.
ffs it's existed since 1997
 
Well, I'm not sure how I'm going to migrate to IMAP4 without breaking local archiving.
 
Bob
   The IDLE command is sent from the client to the server when the
   client is ready to accept unsolicited mailbox update messages.  The
   server requests a response to the IDLE command using the continuation
   ("+") response.  The IDLE command remains active until the client
   responds to the continuation, and as long as an IDLE command is
   active, the server is now free to send untagged EXISTS, EXPUNGE, and
   other messages at any time.
@bwDraco Local archiving depends on the mail client, in all cases.
POP is just... ugh.
 
I'm just not in a mood to change my email workflow. It works for me, after all :P
 
Bob
Eh, your choice.
I know I'd be (hand have been) endlessly pissed off by unsynced read state across devices..
 
Bob
7:13 AM
@JourneymanGeek I... wouldn't mind having that :P
 
@Bob I'm trying to port .NET kynnaugh to .NET Core using Visual Studio for Mac :D
>
using D3F00FF1770DED978EC774BA389F2DC901F4.B00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.C00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.D00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.E00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.F00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000;
I kid you not, that's in the code, and I have no fucking idea why
 
@Bob a raspi to control a CD rom drive tho?
 
Bob
@allquixotic if you actually want to use it, I have a Core port here
and it compiles
it's just missing the resampler
@allquixotic That's part of DllExport. FYI, you can't use that outside Windows.
It relies on mixed-mode assemblies.
With .NET Core, you actually have to write the C bindings.
 
@Bob why doesn't what you have, work?
@Bob ...huh; well I have it building to a bunch of .net DLLs o_O
 
Bob
@allquixotic relies on Win32 DSO API
 
7:24 AM
libgrpc_csharp_ext.x64.dylib
well that's taken care of
 
Bob
I mentioned I was going to switch to libresample, but only if you needed it cause it's nontrivial
 
@Bob is that CSCore?
 
Bob
@allquixotic but you won't be able to load as plugin without the native binding
@allquixotic yea, CSCore uses it
there's no pure .net replacement
@allquixotic heh, grpc is properly multi :P
I'll look when I get home
in an hour
 
@Bob are you sure? it seems to be using ffmpeg under the hood on other platforms.
it seems to claim that certain codecs are only available on Windows, but FLAC and WAV are cross-platform?
 
Bob
@allquixotic ...possibly, but I looked at the resampler source
also cscore flac is decode only
using flacbox to encode
 
7:29 AM
hence flacbox. ah
well if I can get it into a dylib, I'm pretty sure it has a chance of working, since I have MSIL .dlls of the entire solution right now
and natives of the bits that need to be native
 
Bob
@allquixotic get what?
oh, just use libresample :P
I was thinking about porting libresample to C#
it's only like 1k lines of math code
 
@Bob I need to basically find something that's .NET Native but for OS X / .NET Core :P
 
Bob
@allquixotic oh, that
@allquixotic I mentioned before how I'd do it for Core
 
that would be awesome
 
Bob
@allquixotic basically, even DllExport doesn't work for Core. You need to make a C lib project and use the .NET Core API
so you manually export the C functions and just load .NET to call the transcription bit
of course that's where I thought about going back to pure C++ -_-
 
Bob
@allquixotic Huh. I forgot about that one. Does it export functions?
NativeCallable
Awesome! :D :D :D
> Apply [NativeCallable] attribute to a managed method and then it can be called from native code.
:D
hmmm
 
delicious
so basically there's no reason to use C++ anymore :D :D :D :D :D :D :D :D :D
.NET will inherit the earth
 
Bob
@allquixotic I'll just ... uh ... stop procrastinating and do the needful (TM) and actually do the cleanup :P
 
@Bob I've already done a fair bit!
should I commit it?
 
Bob
@allquixotic oh, nice
is it pushed?
 
7:42 AM
no
I'm afraid to because I think I broke some things
 
Bob
ah
@allquixotic might as well push :P
though if we're going to Core there's not much point doing more on the Windows-specific one
 
I think what we have isn't even Windows-specific, we should be good
 
Bob
@allquixotic Except CSCore and DllExport
the rest is platform-agnostic managed code, yea
 
I don't think CSCore is going to be a problem, but I'll test it.
 
Bob
@allquixotic I'll be amazed (and very very happy) if it works
spent a whole day looking for alternatives
@allquixotic Oh, it was DMO. github.com/filoe/cscore/blob/…
 
7:45 AM
uhh, I think NativeCallable is the other way :S
 
Bob
@allquixotic nah. that's just the primary use case, to provide a callback
see comments on that issue
> Please correct me if I'm wrong, but NativeCallable will allow me to export my class library so they can be called from my native application as if it was a native dll (think LoadLibrary then GetProcAddress), providing I run it through your AOT compiler and I link MTR.dll? What's the catch? Is this only going to be for WinRT?
> Yes, NativeCallable will allow you to do this. The catch is that it will take a while until the CoreRT compiler and frameworks are complete and production quality (we do not have date yet).
 
pushed my awful changes
we'll revive the unit tests later, I was just lazy
@Bob what issue?
 
Bob
@allquixotic I was going to merge the projects anyway
 
@Bob right, that's one of the things you mentioned so I did that
 
Bob
13 mins ago, by Bob
https://github.com/dotnet/corert/issues/768
train just got to station, so I'll be walking. back in 20-30 mins
 

« first day (2345 days earlier)      last day (2681 days later) »