Fork me on GitHub
#clojure-uk
<
2019-03-04
>
guy08:03:41

morning!

thomas09:03:46

is it just me or is programming with @annotation (like the Java ones) actually pretty crap? It requires a good understanding about what the annotation means, what it does and how it is affected by other annotations. It isn't code, and that is not a good thing (I think) as it seem to complect things.

šŸ˜‚ 5
šŸ’Æ 5
šŸ’© 5
thomas09:03:10

example: we had a performance problem with the DB... developer adds some Hibernate Annotations... performance improves... but other things break. suddenly not all the fields from a class get serialised anymore into JSON by Spring due to some other annotation.

šŸ˜± 4
thomas09:03:53

I don't want a change to a setting (what an annotation is to a certain degree) to also change the behaviour of my code!!!

thomas09:03:21

just me? or am I barking up the wrong tree here?

Ben Hammond09:03:48

It was my experience that Hibernate was very good at the trivial stuff, but very bad as soon as you wanted anything more complicated than 1Row:1Object

Ben Hammond09:03:28

Depending on how your database <-> object mapping are it might save you alot of time

Ben Hammond09:03:37

or make thing a whole lot worse

Ben Hammond09:03:06

devs get into a pure computerscience mindset whereby you want to solve the problem within Hibernate

Ben Hammond09:03:31

whereas its is often simpler to do outside hibernate

thomas09:03:48

that is my feeling as well... very easy to get started. When you are just starting a new project and the data model is changing a lot it saves you a lot of time as you only need to write it ones (just the Java code, not also the matching SQL). That makes it easy... but in the long run not simple.

Ben Hammond09:03:56

because you are trying to glue two things together that are not quite compatible https://en.wikipedia.org/wiki/Maxibon

šŸ˜‚ 4
thomas09:03:56

as you are comlecting things...the Object model and the relational model.

thomas09:03:30

hmm I quite like those!

Ben Hammond09:03:50

I hadn't realised that the icecreams were the same flavour

Ben Hammond09:03:06

I thought the two halves were wholly incompatible

thomas09:03:38

no, one kind of ice cream... just one side covered in chocolate and the other in a biscuit.

Ben Hammond09:03:55

somehow that is disappointing

thomas09:03:04

still, I don't like Hibernate... too much black magic for me.

alexlynham09:03:36

man ORMs are the worst

alexlynham09:03:13

Rails wasnā€™t even that bad a framework once you ditched the ORM

agile_geek09:03:30

Trivia for the day. In the early days of support for JPA I patch a problem with the cascade delete in Hibernate. It's not black magic but at that time (about 2004-5 from memory) there was a lot of Aspect Oriented code that had to be wrangled if you wanted to understand the guts of what it did. I was writing a load of AOP code (AspectJ) that sat around Hibernate altering what it did for my specific domain. It was a crazy idea that I objected to to start with, cost me a year of my life and, as I predicted, never went live!

agile_geek09:03:39

I tend to sum all of this stuff up with favouring explicit code over implicit means you take your pain up front but it's amortised over the lifetime of your code base, versus easy start and then pain or even a deadend later.

šŸ‘ 16
flefik09:03:00

itā€™s also never that much pain. Itā€™s a couple of extra lines to write. 10 minutes here, 10 minutes there.

alexlynham09:03:36

yeah explicit is literally always better

šŸ‘ 10
maleghast10:03:14

Morning šŸ™‚

Ben Hammond11:03:52

other news sources may be availablel

otfrom12:03:41

starting fires in the great beyond

otfrom15:03:01

I'm thinking about going to Spitalfields Market in London for dinner at around 630ish. Anyone around fancy going? (no idea where there, but lots to choose from)

Ben Hammond15:03:23

I'm not in London I've always eyed the vietnamese Pho place that is round there though

Ben Hammond15:03:41

Not actually managed to get through the door

Ben Hammond15:03:36

(not sure what time they close)

maleghast15:03:14

@otfrom - I'd love to meet you for dinner

otfrom15:03:17

if that is called Pho, I've been to some others before and they are good

maleghast15:03:39

Where are you thinking of eating?

otfrom15:03:39

do time and location work for you?

otfrom15:03:56

idk, that's why I was aiming at Spitalfields Market. šŸ˜‰

maleghast15:03:01

Oh aye - I can walk to Spitalfields from the office

otfrom15:03:26

just need a time and a place to meet and we can decide from there unless you have strong opinions

maleghast15:03:28

There are many chooses there

otfrom15:03:34

that was the plan

maleghast15:03:07

Perhaps we meet by the sculpture on the corner (is it a pear?) at 1830h?

maleghast15:03:47

meet there ^^ ?

otfrom15:03:19

which corner is that?

otfrom15:03:26

nw/ne/sw/se?

otfrom15:03:06

@maleghast I don't actually know where that statue is šŸ˜Š

otfrom15:03:44

@maleghast how about the tented bit by the white goat statue? https://goo.gl/maps/6N8aecVuE8M2

otfrom15:03:59

(near cycle surgery)

maleghast15:03:18

Sure šŸ™‚

maleghast15:03:32

See you there šŸ™‚

otfrom15:03:57

ah, they are very close to each other I see now šŸ™‚

otfrom15:03:52

@maleghast 6:30PM work for you?

otfrom15:03:16

and if anyone else is around and wants to come along

maleghast15:03:17

Indeed - 1830h is great šŸ™‚

alexlynham15:03:28

I would join but my train home is at 1900

alexlynham15:03:36

maybe some other time tho šŸ˜„

otfrom15:03:50

drat. Sorry to miss you @alex.lynham. Just as I'm becoming sociable again too

alexlynham16:03:48

haha there will be other times no doubt

alexlynham16:03:43

also in case there are any peter gabriel fans that werenā€™t aware (or in my case, werenā€™t born yet) - holy s**t the german versions of Melt and Security are mental

jasonbell16:03:11

@alex.lynham theyā€™re brilliant, I had the on vinyl years ago

jasonbell16:03:17

PG2 is still a wonder.

alexlynham16:03:01

currently listening to Schock den Affen šŸ™ˆ

otfrom16:03:25

oh Affeeeeeeeen

jasonbell16:03:48

Warning: Contains Chapman Stick

Conor16:03:27

Oh, the Sledgehammer guy

Conor16:03:53

Didn't he turn into the T-1000 in the video for that

jasonbell16:03:21

Noā€¦. that was Brian Pern, who also had the lowest recorded bass note in pop music.

Conor16:03:03

You need to listen to some nice German rap instead

Conor16:03:16

Although frankly Krieger is the only Fantastischen Vier track I've heard that I like

jasonbell16:03:10

@alex.lynham Still prefer PG3 when itā€™s all said and done. Because thatā€™s where I Donā€™t Remember Isā€¦..

otfrom16:03:24

sorry, I don't recall

trollface 5
jasonbell16:03:22

oh well played

jasonbell16:03:36

Ah well, moved on to Toni Childs instead.

jasonbell16:03:58

Just on the hope that @otfrom doesnā€™t know any lyrics or tunes.

otfrom16:03:05

hey, don't you walk away @jasonbell

jasonbell16:03:06

@otfrom Did you ever listen to the last St Vincent album, just vocal and piano? Stunning it is.

otfrom16:03:30

something post the best breakup album ever?

otfrom17:03:35

if anyone else is up for dinner in London this evening, @maleghast and I will be meeting at Spitalfields Market at 18:30

flefik18:03:30

spitalfields has so much good food! enjoy!

3Jane20:03:58

Shame I missed the impromptu meetup, my phone died :(

3Jane20:03:43

Hereā€™s a thing for tomorrow, relevant to what we talked some time ago (front end, gendering of tech, and the effect of boot camps.) https://www.theguardian.com/technology/2017/mar/14/tech-women-code-workshops-developer-jobs

rickmoynihan23:03:52

A thought provoking piece; though I think it ignores the reasons coding initiatives tend to teach front end not backend skills; which is that front end gives you direct visual feedback, and a feeling of creating something sooner. I think this is what biases entry level jobs to those with less experience. Itā€™s not that itā€™s harder or easier than backend; itā€™s just easier to get people interested in coding when they can see something. Thereā€™s more reward and sense of achievement earlier on. Where as I think backend relies much more on the thrill of solving a problem rather than creating something tangible. On a more positive note ā€” I get the feeling (from purely anecdotal evidence), that more women might be entering data-science, than other areas of computing - and that is comparatively higher paid than other areas. I have no stats to back this up though.

3Jane06:03:49

Yeah, this is also my impression, that we get more PhDs who want to be able to make a living and not work insane hours. They leak in from other disciplines.

3Jane06:03:18

Also agree with things being most accessible being most denigrated at the same time.

3Jane06:03:48

Iā€™ve been thinking on how to give learners the best feedback and feeling of accomplishment and you know what? Excel.

3Jane06:03:41

Now think of how many people youā€™ve seen sneering at business procedures encoded in Excel, and how they should bring in true programmers and make it all better.

rickmoynihan09:03:09

> Also agree with things being most accessible being most denigrated Iā€™m not sure theyā€™re denigrated as such; just if itā€™s accessible and entry-level thereā€™s more competition for jobs and less for skills which lowers wages. Without meaning to be disparaging of learn to code bootcamps - but I think people think of theyā€™ll ā€œget rich quickā€. Being realistic you shouldnā€™t expect to learn anything valuable to a high standard of proficiency in a few months. A more realistic expectation is to ā€œget rich slowā€, and spend a lifetime learning. If youā€™re getting into a new career with only a few months of experience it seems reasonable that you might expect to start at an entry level. Unfortunately this all starts to sound like the meritocracy it is. I read literally all those links Sean posted on that subject; but I didnā€™t see many proposed solutions to thatā€¦ just lots of justified complaints and problems about it.

rickmoynihan09:03:08

I agree with all your comments about Excel too. Excel sets a pretty high bar to compete with.

3Jane20:03:06

Iā€™m glad people are starting to question the ā€œletā€™s teach everyone to codeā€, because realistically once everyone is a coder, coding will no longer pay and weā€™ll be back to the starting situation.

guy20:03:41

I find their example salaries a bit iffy

guy20:03:45

Devops vs front end?

guy20:03:50

šŸ‘€

guy20:03:19

the difference between software engineer and front engine is 37k to 33k

3Jane20:03:48

Sorry, that website (glassdoor?) is unusable on mobile, canā€™t tell :(

3Jane20:03:41

Salaries can wildly differ depending on location and also whether the company thinks can get away with it.

guy20:03:02

yeah no worries, i just am not sure why those chose devops haha

3Jane20:03:43

The author isnā€™t a developer and might have chosen that as a representative of ā€œbackendā€.

šŸ‘ 4
3Jane20:03:47

Dunno :) The stats I found on gender spread came from surveys ran by websites and magazines aimed at engineers, so that could have also been somewhat self selected.

3Jane20:03:32

Iā€™ve always wanted to know more about what the industry Really looks like, you know? Because we must be missing huge swathes of population that is not engaged in the same way

šŸ‘ 5
seancorfield20:03:56

"The pipeline is not the problem; the meritocracy is the problem." <-- this.

šŸ‘ 5
3Jane22:03:06

I thought the author said the (lack of actual) meritocracy? As in peopleā€™s tendency to assume thereā€™s no smoke without fire

guy20:03:17

:thinking_face: what do you mean

seancorfield20:03:51

There's plenty of material out there @guy that explains why the so-called "meritocracy" is a problem. A good place to start is http://geekfeminism.wikia.com/wiki/Meritocracy

guy20:03:52

Forgive my ignorance but interestingly Iā€™ve never thought of a meritocracy as a negative thing. Now I can start to understand why it isnā€™t as fair as I first thought! So thanks for this, really helpful šŸ™‚

guy20:03:30

nice one! Iā€™ll check it out

seancorfield20:03:39

It speaks to a problematic mindset in Silicon Valley as a whole (and it's part of why I detest SV "culture").

šŸ‘ 5
flefik21:03:04

devops vs frontend salaries isnā€™t a fair comparison, but surely on average backend developers should command a significantly higher salary than their front-end counterparts? There are whole classes of problems that a front-end developer never needs to care or know about. JS development is literally just single-threaded CRUD and positioning of elements on the screen. Thereā€™s no scaling issues, no race conditions, no out of hours support rotas (urk), no data migrations no exploding databases etc etc etc. It really is a lot simpler than backend work. Am I wrong about this?

Conor21:03:48

Yeah dude, probably. Frontend salaries are lower because the supply is higher, as far as I can tell.

alexlynham22:03:37

ho ho ho ho ho

alexlynham22:03:18

try and build a scalable SPA that works across devices and then tell me how terraform and jenkins and bash are harder šŸ™‚

seancorfield22:03:43

As a backend developer of 35-ish years, I certainly do not begrudge my frontend counterparts their high salaries, considering the crazy minefields that frontend development is. Cross-browser behavior is a nightmare, and the whole toolchain shifts so rapidly it's a full-time job just keeping up (we're just going through the WebPack 4 pain right now).

alexlynham22:03:46

Iā€™ve often found front end or JS folks are as good if not better, they just donā€™t always have much theory (just my experience, sample size of 1) - also true that two of the best engineers I ever worked with were JS

seancorfield22:03:45

You couldn't pay me enough to take on frontend work. Sure, the problems are different between the front and back and FE folks often have zero idea about the weird stuff us BE folks have to deal with -- but the opposite is also true.

alexlynham22:03:19

in my last work I was nominally responsible for a team where a female engineer who did full stack JS and was the best dev on the team was routinely sidelined by a male colleague that was just a web designer with some JS

alexlynham22:03:57

trying to put her in the driving seat - she was smarter and more professional was difficult because of the biases of people around her

alexlynham22:03:50

even though it was clear she was the best (to me, anyway) - wouldnā€™t have surprised me if she was paid less too but I donā€™t know what her comp was

flefik22:03:09

i have seen this many times as well. itā€™s almost like in nearly all organisations there exists a senior (by time with company, not skills) in sidelining a promising junior (again, by time with company).

seancorfield22:03:00

We get to test our stuff in isolation or in carefully controlled environments. They can't even begin to test in the actual environments their code will run in. I am continually horrified at the problems that they have to debug and resolve based on bizarre outliers caused by browser/plugin/OS combinations out there šŸ¤Æ

3Jane22:03:43

All of that ^ by Alex and Sean really. Plus, not all backend work is complex.

3Jane22:03:49

Itā€™s been my general experience that when I think ā€œX is simpleā€, itā€™s just that I didnā€™t know how complex it was because the pros doing X handled the complexity for me. And when I learn more about X, my mind is blown. (Most recent reason: marketing.)

3Jane22:03:24

And you couldnā€™t pay me to do frontend (people tried, I wonā€™t, I still have IE6 scars.)

flefik22:03:46

thanks for this. iā€™ll need to ponder a bit more, but donā€™t think youā€™ve quite convinced me yet. Iā€™ve done both frontend and backend for many years, and I still feel that the breadth of knowledge and skills required to be successful on the backend is much larger. That said, @seancorfields comment about ā€˜carfeully controlled environmentsā€™ resonated with me.

alexlynham22:03:23

> when I think ā€œX is simpleā€, itā€™s just that I didnā€™t know how complex it was because the pros doing X handled the complexity for me. oh man this 100%

alexlynham22:03:44

project, delivery, user research, you name it

Conor22:03:36

Everything is fractal