rlayton2

My partner is the same. I have a document that says "here is the name of our accountant. I recommend you do X for now".

Thanks for that.

A related note is for any requirements files. Something like this bit me the other day.

Libraryname >=3.1

After a few years,the package was updated substantially and has lots of breaking changes in the recent branch. Fix was to so ==3.1 until we work out the next step

Rejected, please don't commit code in a broken state.

Similar to my strategy. Write the code as if I had all these magic functions that did steps:

    def write_report():
        db = connect_to_database()
        data = get_data(db)
        summary = summarise_data(data)
        save_report(summary)

After this, do the same for each of these functions.

Total layer size is also restricted.

Which I've had failed as a student in Australia, as we use .edu.au (not for Dropbox, but other services).

As you said though, its a simple test, and if you don't think about it too much, its too easy to just test the email ends in .edu and move onto the next task.

Agree, and nothing tells an employee they aren't valued more than not providing them an adequate workspace

Ultimately, if you are using the smartphone for a deliberate purpose, that's probably fine. Looking up a recipe, contacting a loved one, reading an ebook.

The problem is when you are just "on the phone", not really doing anything, not anything you planned to do, and not really getting anything done. Even this can be fine, as long as you go "I'd like to see whats on twitter for half an hour", rather than just opening things up and doing them because the algorithm pulls you in.

Yes, but many non-technical business owners want to focus on other aspects of their company and just get someone to do that work.

Legacy work includes websites that were written a few years ago, but still work and don't need to be completely redone. Add some features, but a rebuild is unnecessary.

Yes! Just don't cold-email businesses with your offerings - lots and lots of low value options do!

Instead, find local meetup groups of businesses and start making connections. Be the local business doing this.

No, but it does apply to the distribution of slippery slopes.

The OP specifically said tools "like this"

Sure, but then that reduces the effectiveness of such techniques by 1/4 or 1/10, while still carrying the risk a secret shopper finds it.

I've found it helpful to deliberately identify if you are being the manager or the technician (to use E-myth's terminology). The manager goes through the lists of tasks and emails, and works out what needs to be done (next, or today, or this week, or this sprint). The manager tells the technician (usually via TODO list) what that task is. Then you switch modes to technician - its no longer your job to organise, its your job to do.

So you have an example? I've never seen this and would like to see what people are doing here. I teach a lot of newbies data mining, so I'm very interested in how people get it wrong

Yeah I thought I was reading the title wrong!

For anyone that isn't aware, the role of pca is to create new (synthetic) features that represent the original features.

It does not tell you which features of the original set are good for feature selection purposes.

For the people selling, it is often more profitable (or at least less hassle) to sell quickly with minimal fuss and move onto the next one.

You are generally right. The data is a big difference. Managing it, it's too big for git etc is a massive problem.

ML also has a distinction between the system for training a model and the system that uses it, but that's just two projects to manage

I think like lots of things... If you expect to do better than other people, be prepared to know more or do more then them. Many people dedicate their lives to learning the stock market and what to do. It's unlikely someone can beat them with doing the bare minimum research.

Not saying there isn't shady stuff going on too. There definitely is, but even if there wasn't it is a skill based game.

It's newly up so you were tied fifth!also go for the best ranking which is closest to 1

I have a small game that I made to experiment with some tech.

https://donotplaythisgame.click/

In short, you accrue points while you aren't playing it, but you have to log into actually get the points (similar to "the game", which you just lost). You get your global ranking only after logging in though.

I like that a few video games are multi-words (e.g. Oxygen Not Included) and use a common abbreviation, (ONI) that is google-able - for instance "oni tutorial" is pretty good, or "oxygen not included tutorial" is a sure thing. Maybe more software needs to have more words in the name.

Those might be the same people who say "why do you do research in field X, rather than on cancer research?". Why do you have a job as an admin rather than go back to school and become a cancer scientist?

It's easier to criticize the actions of others than to take responsibility.

This controversy will be over in a few months at worse. Spotify nullifies the damage by removing the music now, and the investment in the music would be a long term investment.

I'm not a big fan of them either. For teaching beginners, its often "look at the help for this function, but ignore these bits for now", which is quite jarring.

They are useful, but I think sometimes the advanced features run into the basic interfaces a bit, making it weirdly tricky to teach beginners without being confused.

Absolutely. I often joke that my work as a data scientist is mostly creating bar graphs for people. The actual analysis is often reasonably simple, its the aggregating of the data that is hard (its messy, its not all in the one spot and there is lots of it).

So start with querying your big data to say what the top three event types are. Then slowly crank up the analysis complexity, but not too much. The data engineering has lots of scope for real solid and obvious applications.

In most cases, the productivity tool is about the habit, not the specific tool. If a text-file todo list doesn't work for you, try something else. However at the end of the day its the habit, not the tool that is the most important thing.

Isn't the normal process, roughly, to:

1) Use unit tests to move quickly through to implementing The Feature 2) When The Feature is complete, run in a local environment to confirm it works 3) When that works, move The Feature to a development environment that more closely mimics Prod 4) Deploy to Prod

Each one of those steps takes an order of magnitude longer than the previous one, so should be done an order of magnitude less often. However if you are finding inconsistencies between steps, then alter local/dev to ensure more consistent testing. (i.e. if it works on Dev, it should just work on Prod, however there are always little issues).

A significant improvement on letter-by-letter, but not that much harder, is to use n-grams: "two letters to predict the third" etc. Still not "industry grade", but the results start making more sense.

I closed my LinkedIn account due to recruiter spam. My account said I wasn't looking for work but I still got lots of messages that were just keyword hunting.

A good recruiter is gold but it's a profession that has a large number of spammy practices.