Notices by Gwmngilfen (gwmngilfen@fosstodon.org), page 41

@iona ooh, do like. I'm a big fan of "shenanigans" too...

@alex oh, and Ringworld by Larry Niven is a classic too

@alex for short-but-packs-a-punch, Flowers For Algernon by Daniel Keyes. A great example of 'what if?' SciFi without jumping to space/future stuff. Packs an emotional punch.

For longer form, anything by Peter F Hamilton, but top pick would be the Commonwealth Saga (Pandora's Star and Judas Unchained). If there's a creature like that out there in our galaxy, we're all doomed ;)

#scifi #MorningLightMountain

@cassolotl great work there! I was also surprised by the x-axis on the satisfaction chart, much closer together than I expected.

I may take up your offer of playing with the raw data - my fledgling #rstats skills need practice.... If there's one more in-depth question you'd analyse from your results, what would it be? I feel like checking the statistical significance of some of the small-group results is worth doing, but the last thing I want to do is step on your toes!

A rare day off today, just myself and my good lady. Not a common occurrence at all, and all the more wonderful for it

#partnership #team #romance

Here's a local river we went walking along...

I mean, I hope I'm overreacting. I hope this means "users can see that a business has WhatsApp via its website or similar, and *choose* to interact with the business that way". And I'm sure that is a part of the offering for businesses.

But I doubt that's the whole of it...

#metadat #socialgraph

@mooshoe indeed. there was a great article I read recently which pointed out that in many cases startups were claiming to use AI (for funding) but then actually employing people to do the work.

For experienced analysts out there, critique is very much welcome. I wish to improve :)

I guess I should probably do some #self #promotion or something, since I''m trying to get more into #data (and it's a good intro to my new instance :P). Blowing my own trumpet doesn't come naturally to us Brits... :)

Still, here's a blog post I published this week on forums vs mailing lists, and what it meant for our community.

https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html

I will say this, learning #rstats has vastly improved my ability to make pretty graphs...

#ggplot2 #regression

@mike sometimes it's ok - I really don't care about the 50 regressors used in a movie recommendation engine. But when I hear about the nonsense happening with ML in, say, healthcare... I worry.

@mike thanks, good to be here. Sysadmin work is usually a thankless task, so kudos for running an instance!

Started my machine learning course this week. It will be interesting to see what the reality is compared to the hype.

As of right now, I'm feeling that ML is what we do when we don't really care about understanding the situation - prediction is the only value. That feels sad to me, as an ex-physicist I want to know *why* an association matters. Linear regession was a ton of fun for exactly that reason - we'll see if ML can win me over :)

#rstats #coursera #data #regression #ml

Alright, time for a new home. Fosstodon has been recommend to me by quite a few people as the place that matches my interests, so, uh, here I am!

For those not following me over from mastodon.org.uk, I'm a FOSS community manager, wannabe data scientist, sysadmin and devops enthusiast. Currently very interested in #rstats and #data #analysis, but will happily talk endlessly about pretty much anything.

#introductions