Hacker News Reader

Comment by brooklyndude | Hacker News Reader

I can hack d3js. Can crunch billions of records in a flash of the eye, can wrangle most any DB (NoSQL, et al) out there, why would I need Tableau?

Answer: We want someone stable! Not some crazy hacker. Plus if we don't spend the cash, we don't have it in our budget next year.

Oh, I get it now. :-)

View on HN

No you don't get it. Tableau is for people who can't hack (or don't want to hack) d3js or crunch billions of records or whatever techie thing you seem to be so proud of. The mission is supposed to be self-service business intelligence, i.e, allowing a typical business analyst to drag drop and create good visualizations and analytics. Not dick around and waste time with databases and javascript.

Except that it only lets you do a set number of things. Most requests for information are not simple, and once you leave the comfort of simplicity, you have to become an expert in tableau, or in many cases, just accept that it can't do what you want it to do.

Unless your queries are very simple, you are better off finding a programmer and asking them to do your job.

Tableau is for business users to generate graphs. Also d3js is a bear to work with until you've steeped your brain in it.

By your logic, there's no need to use MS Word, people should just use TeX.

Well said, the word/tex analogy is spot on.

Lol, so I use Tableau every week. For example today I had a government request to measure quantity of customers outside a particular business radius of 40km. Meaning what percent of total visits are in the radius vs outside it. I had 8k postal codes.

This took me literally 5 minutes in tableau, I would hate to see how long this would take you writing code.

..Just the tip of the example iceberg.

A programmers whose job is to make these kinds of reports will already have information about postcodes and so on, and take less time than you. Tableau is quick at doing simple things, but takes a lot of learning to do complex things. Often, you will have to resort to custom scripts, creating views, etc. Often, it just won't have the ability to do certain things.

Anyone with any experience in drag and drop applications will tell you that once things get complex, it ends up taking much more time using the simple drag and drop solution as you spend all of your time trying to get it to do something that is outside of a basic drag and drop model.

Are the complex requests in regards to getting/manipulating the data, or in the final visualization. I'm trying to build a desktop tool similar to Tableau but with the ability to easily drop down into python and do data manipulation with the pydata stack. Intended audience is the new breed of quantitative analysts/data scientists.

I would say mainly visuals, think Photoshop meets Excel meets Javascript. One of the founders of Tableau founded Pixar studios so you can imagine that graphics are important. But it also has some great data options.

Some examples: https://public.tableau.com/s/gallery

Both. While tableau has a visual query builder, sometimes the queries are too complex. For example they may need self-joins, non-standard join criteria, temporary tables, etc. So ultimately, the tableau guy has to ask a programmer or database person to create a script to use as the tableau source.

One example that involves both is when the visualisation needs a contiguous date range in the data, but the data is missing some dates. As a programmer it is easy to just loop through a date range and put 0 where there is no data.

There are also lots of limitations in terms of the visualisations. The end result was that multiple visualisations had to be created to show something that a programmer would be able to create as one visualisation.

You say "will already have information about postcodes and so on". How exactly do you get your hands on postal code geo data, have you tried that? Have you created your own geo postal code database, or maybe used an API. I'm talking international postal codes too.

I have, and I can tell you it's not easy, for example in Canada there are like over 10k postal code regions. Then imagine maintaining this dataset, no thanks.

Anyhow that's not the point, that's just one data set Tableau does very fast, but tomorrow could be something totally different and the next day, etc.

Sure for complex custom requirements no BI tool will work, but it's sure better than what people were doing just a few years ago with excel or hiring expensive firms.

Seriously? It's almost as if companies don't want to hire a badass hacker to generate a business report used by non-tech workers. Tableau has a huge place in the BI community, and the last thing that a manager wants is somebody hacking d3js to to create some visualizations.

$1,000 to generate a business report? That's how much a single Tableau license is going to cost you. You still need to train up the staff member to know how to use it (yes, it's probably the most intuitive BI tool on the market, but like anything it has a learning curve).

That's a lot of money for a business report.

Do you pay tableau $1000 per business report?

Nope. But if you work out costs, I think you'll find a dedicated developer on a contract may well work out the same amount as you spend on licensing.

No, you won't.

1,000 USD/yr for a license.. If you pay someone 100 USD/hr to do the work compared to 50 USD/hr for a random Joe Blow analyst then it only needs to save 20 hours of time.

ITT: A lot of people who don't realize 1k a year isn't an absurd enterprise price.

Also to clarify, the licence is $1000 for the first purchase. Subsequently, you only have to pay the maintenance fees of about usd 200 per year to remain eligible for support and upgrades. You can still continue using the latest version you were on in perpetuity.

How much for licensing the server itself?

I don't know any of the server costs. We pull our data off our own servers and handle all of that ourselves.

I can do all that too, but I'm fairly sure experienced Tableau people can do it significantly faster than I can for a comparable quality of result. (Tableau is the only non-open-source software in the stack my company runs).

Actually the analytics is the missing part in d3 and charting libraries. http://www.infocaptor.com/dashboard/charting-engines-vs-anal...

Yup! Tableau isn't just about the end visualization. It is about the exploration of your data and the analytics that go behind the final visual. That is where the power of Tableau lies.

You are 100% correct. I worked in a company that used Tableau heavily and I am pretty sure I know more about Tableau than most people here.

or plotly ... or shiny ...

https://plot.ly/javascript/ http://shiny.rstudio.com/gallery/

Reply on news.ycombinator.com