4ae23193
extracted
15. Lightning Talks 2 - wroc_love.rb 2024.lightning.txt399b6c6cc319| Status | Model | Tokens (in/out) | Duration | Cost | Nodes/edges | Read set (nodes/edges) | Time |
|---|---|---|---|---|---|---|---|
| completed | claude-opus-4-7 |
139,120
/
4,592
30,010 cached ยท 6,002 write
|
78.1s | - | 7 / 14 | 69 / 0 | 2026-04-18 06:48 |
| failed | claude-opus-4-7 |
NoMethodError: undefined method 'with_indifferent_access' for an instance of String | 2026-04-17 23:20 | ||||
| failed | claude-opus-4-7 |
RubyLLM::BadRequestError: You have reached your specified API usage limits. You will regain access on 2... | 2026-04-17 16:18 | ||||
so uh this is a Libra office default
template this this cool effect and um
I'm Michael from liik and I thought
before yeah before letting the lightning
TS die I talk or talk briefly about Zite
which is a um open source race
application uh which we've been um yeah
like working on for quite a while as a
side project
um so I copied this from our website uh
basically Zite is kind of U something in
between tro and like a big Enterprise
solution from
Salesforce uh think like task and
Process Management and flexibility and
inputting data and getting structured
data out and so on
um
but it's only available in German uh for
now and uh so would I would I advise you
to to to use it um probably not because
also it's not meant for you as you can
read it's uh I mean like people like you
and me we have other means to set up our
like don't know task tracking and I mean
who who who invented issue management
yeah um so kind of uh it's not not
really that that useful for us but uh
for for like our C or not customers but
our users it's it is pretty pretty
useful so um and it's open source uh and
we're really happy if you have like if
you if you ever get the idea we we need
something in between like this ex
existing Solutions and should be open
source and maybe gdpr stuff and so
counts uh then just like contact us um
we also very happy to uh implement it
into like research projects it's what
what we're actually using it uh uh most
for um but why I'm am I here uh so
recently or also the last two days we
heard a lot of like projects with which
which have a big Outreach and yeah kind
of like stars on GitHub and G lab they
count quite a much quite quite a lot so
and we we really had a had a major surge
in popularity recently um
50% uh turns out yeah our our users are
not really on good good lab yeah um and
every time I I tell somebody here on a
conference and tell them hey and why not
give us a star he's like oh yeah I will
do so but then there's two Factor
authentification and it's just too much
too much work so I would be really
grateful if we could like up the num for
like maybe reach 100% growth today or
tomorrow um that would be
cool um
yeah and then I thought
like
sorry oh
great I have to make new slides now yeah
um and then I thought about maybe like
some something completely um unrelated
which I could maybe talk about also uh
we recently or like a while ago had a
had a new research prototype which we uh
should like had a like a contract would
implement it no designer and no no real
UI requirements everybody saying UI
doesn't really is not so important it's
the first prototype and then we'll see
uh so and then I mean you can spend days
just like researching which option to
bundle your CSS and JavaScript and the
world is changing and do you want to
take tnd and da UI or bootstrap or
whatever so we really um sat together
and made a made a decision um to choose
nothing uh
so and it's it's kind of it was it's was
interesting yeah because it takes you
quite a for quite a reaches quite a
while yeah it's a uh so pure HTML it's
kind of nice it's mobile friendly yeah
more more than mobile friendly maybe and
um but of course at some point like the
users uh they were like okay yeah it's
everything is nice I mean you can have
like this multi select uh Fields like
like this is example I mean you have to
know how to select multiple items in it
but it's works out yeah um and then they
came up with yeah but really I this are
we have this big data sets of documents
and we really we need to get a get a
idea of of the distribution of the data
like how many documents per year maybe a
histogram or something similar and I was
ah okay this means I don't know like
some kind of spark line inline image
live I don't know yeah
but yeah then I really I I really didn't
want to include any more like big UI
libraries for this phase of the project
um and I don't know if anybody else here
has an idea what how we solved this uh
the solution was actually I think was
kind of a nice uh getting the job done
um because we simply uh used Unicode
spark lines which is kind of a thing as
soon as you
know how to Google it it's there like a
huge page of how to implement unic code
spark lines on on some Viki um yeah and
it just worked out perfectly and they
were happy we could really nicely uh
give them a feeling for the data um for
not only selecting the data sets but
also uh for results and um yeah that's
all I go back to the uh
thanks