3amOpsGuy

Great tutorial, a good level of detail too.

The only other thing i'd mention for Flask newbies would be the excellent debugger facility in Werkzeug (effectively a part of Flask).

I lied, one more reference - the guy that wrote flask (bit of a genius, also has a bunch of other excellent libraries) has shared slides from some talks he's given, they're really worth a read too: http://lucumr.pocoo.org/talks/

Bucket list 2013, attend a talk by Armin Ronacher.

GAE's working very well for me, i use flask on GAE.

I'm still bemused at how much of a free quota you get on GAE. I've had days with a few thousand hits, that doesn't even take any of my daily quotas above 35%.

2012 has been an awesome year. UK based here so the Olympics were a high point (despite beforehand being fairly against the "waste" of money - i was wrong).

This is seriously cool.

I keep mixing up Eben Moglen and Elon Musk, am i alone?

Thanks all for the leads. Hetzner are seriously low priced, impressive.

In the end we've gone for co-location, we tried to avoid this due to the hassles we've had in the past around physical access.

I guess this workload isn't a great fit for the cloud just yet. Maybe some day!

Valid point and I fall into this camp for all sorts of key software (e.g. you'd never catch me spending more than 5 mins tweaking a new web browser).

For other software, e.g. my editor, my mail client - software i'm producing content in, rather than consuming, i'll happily put in a few hours up front to get comfortable on the grounds that i will reap a greater reward in time saved medium term.

The idea of running a desktop "in the cloud" on EC2 or similar interests me. Not sure how practical it would be in real life though.

Been through the same too, it was about your age at the time too - wonder if that's a common theme.

I fixed it with a change of direction. I'd been tackling a bunch of stuff that my heart wasn't in, figured out what i did still like (turns out i wasnt doing any of it day to day), engineered a change of circumstances at work and popped out the other end feeling happy again.

End to end about 4 months. Did involve a fair amount of persuasion - what i wanted to do didn't fit with the organisation at the time.

Get rid of the .json, use a specific media type, if at all possible refer to a schema description in the metadata presented.

Search is a service, not a part of a resource. Whether a seperate search service aggregating all different REST resources is supplied or not is irrelevant in the example case as resource collections can be filtered, which the article highlights (the colour= parameter), for a dogs resource, or a phonebook resource etc. filtering on attributes (which can use wildcards or regexes as seen fit to do so) provides the necessary functionality anyway. Search is a diversion in these cases.

Go versionless.

.json in the resource identifier? "Verbs are bad", except when it's "search"? /v1/dogs/molly vs. /v2/dogs/molly ?

Colour me unconvinced! I don't really see how promoting this hacky form of REST moves us forward.

>> though none of their plans have been included in the spreadsheet.

Which is a shame because between Voda (work phone), O2 (previous long term customer) and Three (past 2 years), Three wins hands down, previously O2 seemed the best - to me anyway.

Cheap as chips too, £15/mo PAYG, unlimited data, 300 mins, 3000 sms. Can reduce to £10/mo with a 12 month contract but that also caps the data. They're tethering friendly too.

I can relate to what he said, like the grandparent here it made me laugh.

I wouldn't say i espouse the view presented (have too much respect for what's been achieved in the other camps such as Ruby and even PHP), but internally yeah i think i do view the world that way.

I don't think it's any different from the common "anyone driving slower than you is a doddler, anyone faster is a maniac" - a patently absurd view to hold.

The enlightening part for me is that other pythonistas think this way, because as you said, i have also never experienced this view expressed publicly by the python community.

Good read. I don't think you were controversial :-)

Spotted a wee typo about 1/2 way down:

>> LRU works by evicting less commonly used data in preference of more frequently used data

For this question:

>> Does anyone know of recognized tools which solve this problem?

BMC's Control-M product manages this fairly easily, although it is easy to let the workflow become unweildy with that product in my experience. AutoSys fairs a little better for this use case.

Open source wise I guess you could use PBS or something of that ilk to replicate. I think though an ideal architecture for this problem wouldn't be what's currently available.

I think a hot-hot message queue with deduplication would be a better approach. You can afford then to have multiple hosts submit an appropriately named job and the first node on the other side of the queue to successfully lease the message wins the right to run the task contained within. If it fails to complete the next node leases the task.

It would require some consideration about ensuring integrity of the message and authentication requirements for publishers.

Your direction sounds pretty much ideal. I'll keep checking in on this project.

Meantime I've just installed go... Wish me luck!

This framework could be enough to tip me into a semi serious go powered side project, which i really must get the finger out and do!

How should deployment work for an app created in this framework?

From the "How Revel Works" page it mentions:

    >> If the app compiled successfully, it runs the app ...<snip>...

Is it simply a case of lifting and dropping the binary (ensuring it's linker dependencies are satisfied on the dest host)?

Would the final artifact be deployable to google app engine?

Incredible endeavour and i wish this catches on in a bigger way.

For now the average person can probably jump on this idea to some extent by switching to a tablet where possible. I think the ipad3 is around 4.8watts over a full charge discharge cycle with the charger unplugged after charging.

I personally feel the parent of my comment is the most balanced view on the page currently.

For all I am liberal, or at least centre left, and a free speech advocate, I cannot see any benefit to the existance of child porn.

I would however like to see a reclassification (reduced scope) of what is and is not child pornography, and also introduce a requirement to prove willingness on the possessors part.

I can happily skip the strict black and white view of the world promoted in some of the top comments in this case. I am completely at ease with the idea of promoting free speech whilst going against free speech in the event of something which is harmful (I am referring to real child pornography not what the law currently classifies) and with no merit, not even the wildest interpretations of educational or artistic merit can redeem the need for child pornography to exist.

Really well thought out article.

The single app vs. multiple apps distinction is a good frame for constructive discussion.

Not so much because Postgres is better or vice versa but because in these days of agile and YAGNI development* there's a real problem with some developers not thinking further than the next iteration.

I like that development shouldn't be weighed down by 'what if we'll need X someday maybe' but it should have at least a rough roadmap - the terminology i use next is irrelevant but I'm getting at the idea of having macro, micro and miso level goals defined - in the future we'll deliver architecture concern Y, so let's not box our implementation into an incompatible state.

I like this but i can't decide if it's technically abuse or not. The paste command will happily parse - (meaning read from stdin) multiple times, so for transposing a list into a table:

  file.txt:
  line1
  line2
  line3
  line4
  line5
  line6

  $ paste - - - < file.txt

  line1 line2 line3
  line4 line5 line6

Combine with the column command for pretty printing. I seem to find a use for this pretty frequently.

I like the simplicity of this one but it's not very useful day to day:

  $ echo *

As a replacement for "ls".

I agree with what you say, but I think we also need to consider cost as well as capability of the underlying product.

Introducing cost massively changes how that equation stacks up.

"How many tps can I run given a budget in the form of these 6 available servers". Well the boxes are under spec for what you want to achieve with an RDBMS, but yeah no problem with XXX nosql product.

>> Facebook dropped Cassandra for it

I guess there's a bit more to it than that. I understand the fb employees that took over were Hadoop guys so it made complete sense for fb to go in that direction.

The bit I don't get is why is hbase still a bit of a turtle. For example a recent paper linked to from here had random read and write workload running at 180k ops / sec on a 6 node Cassandra cluster but only 20k ops / sec for the same workload on hbase.

Yeah Cassandra has CQL and it works pretty well, it's comfortable to use where the cassandra-cli is quite painful for anything other than the very beginning of a project.

I believe in earlier releases CQL was slower but now it seems fine. Also the most popular Cassandra client, hector, supports CQL, adopting CQL should be possible for a lot of Cassandra use cases.

Wow, it's from Fabrice Bellard of QEMU and ffmpeg fame, excellent.

Just so we're all clear, opening any to any rules on the DMZ means no longer having a DMZ.

That could be a valid move - there are always trade offs to consider and possibly the trade offs stack up such that you should no longer have a separate network for public facing services.

However it's part of the ops team's role to mediate all interested parties requirements and not just allow random changes to make something work right this second because one party needs it for their requirements. Every other team might be a bit miffed if we opened that thus breaking our SOX and PCI requirements and so landing us with a heavy fine, or worse.

I think the point you're trying to make is there are cases where such an action could be warranted and so the goal should be to make that resolution process as fast and as painless as possible for all parties.

I think this is a more of a case of wrong tool for the job. For example if i have a fairly static mail server configuration then it could best served by stable - I get security updates back ported but I won't have significant changes in the underlying platform whipping the carpet from under my feet.

Choosing stable for say the website may not be such a good fit though - the website is permanently in a state of flux due to having money and developer time invested with a lot more vigor, introducing new technologies to the stack, I'd be better deploying testing or more likely unstable under that stack.

For all I love Debian, the naming scheme has cost me many hours explanation to senior ops management over my younger years!

All that said, there is also the valid ops concern about developers introducing unproven tech just because it's in fashion even though there is no business, operational, architectural or even development advantage to be demonstrated or even hypothesised from it.

Its not fun for anyone when you have A team of developers with their dummy's out the pram because you wont open any to any routes on the dmz firewall so they can use some new sexy message queue when any one of the well behaved, well proven, stable alternatives would offer the same advantages.

You should put your talent to good use, I bet you have it in you to write a first rate thriller.

(I'm not being viscicious - plenty of people have imagination, plenty of other people can rationalise about what could work in practice, drawing these together is a skill).

Strace is excellent for anything crossing the syscall boundary. Often though, we want to see interactions across library boundaries.

A handy addition to the warrior's toolkit: http://linux.die.net/man/1/ltrace !

>> which lags even end-of-lifed e-mail clients like

>> Eudora for composing, reading, filtering, archiving,

>> and searching.

These words hit the mark with me. I've never found a polite way to word it so I just bite my tongue and avoid being the negative nelly but whenever someone shows me some cool new tool or website, odds on bet its missing a ton of basic functionality that I already had in the boring old pre-web 2.0 version of whatever the tool is.

We're very forgiving of this in general as consumers but it does suck a fair bit.

Does this mean we now have a convenient way to speak JMX from python? That would be very handy!

I'm impressed! Works better than i expected. Excellent, thank you for sharing.

I miss Sun.

They made some seriously cool stuff, and in their later years they were really quite progressive in their outlook and approach.

Jonathan Schwartz, for me anyway, kicked off the whole cool, open and chatty top exec thing that's so common now.