Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Is Tableau Dead?

mergeyourdata.com

Post reply on HN
Print view View original

101–110 of 128 posts

Re: Is Tableau Dead?

#101

Post by dkersten »

DanielHB wrote: post #96
infecto wrote: post #77

Earlier quoted context omitted.

I am not surprised. There are a handful of competitors but I believe the ship has sailed. I am a complete AWS dunce but it's too easy for me to spin up an ECS with spot machines and run my python app on it. When Heroku hit the scene and for many years, there was no easy alternative. You had to go and provision your own machines to then setup.

Is it just me or ECS is relatively complicated? I mean, it is no k8, but still you need to set up a lot of stuff to get it to work.

I actually gave up on AWS forever when trying to use ECS. It was incredibly complex and it kept failing to deploy, but cloudeatch only contained log entries for it about 20% of the time. All I could do was click redeploy over and over — sometimes there were logs, usually not. And they charged me for every single deploy. After a day and a $100 bill later, I gave up and used DigitalOcean’s hosted docker instead. I had it up she’d rubbing in under ten minutes and no bill other than the actual usage costs.

Now I avoid AWS as much as I can. Of course my employers tend to be on AWS so sadly I can’t completely avoid it, but I do my best and would never recommend it anymore.

Re: Is Tableau Dead?

#102

Post by prepend »

Rastonbury wrote: post #43
kimi wrote: post #7

I don't know jack about Tableau, but if the alternative is Power BI, I got a feeling they may be safe. My current company sells telephony and call-center analytics for MS Teams using our software, and during pre-sales meeting, one of the staple questions is "So we can avoid Power BI entirely?" followed by a minor thanksgiving ceremony and presentation of credit card. I get a feeling it's not too dear to the hearts of…

If you haven't used PowerBI recently, let me tell you it is better than Tableau. In my previous work, using PowerBI I can easily drag and drop dimensions and it can suggest data visualization in browser. Now I have tableau and I have to open the laggy Tableau Creator, wait for a sync, then build a visualization from scratch. Maybe its poorly implemented at the new place, but I avoid it when I can - its easier and fas…

This is so interesting as my experience with PowerBI is that it’s just not as good. Not just that the figures look worse and have worse default colors, but functionality-wise it’s just more of a hassle.

What’s nice is that it’s better than Excel so if you want to replace Excel, PBI is great. But if you want a functional data viz that you’re not embarrassed to put on your web site, I’m not sure PBI works.

Re: Is Tableau Dead?

#103

Post by infecto »

DanielHB wrote: post #96
infecto wrote: post #77

Earlier quoted context omitted.

I am not surprised. There are a handful of competitors but I believe the ship has sailed. I am a complete AWS dunce but it's too easy for me to spin up an ECS with spot machines and run my python app on it. When Heroku hit the scene and for many years, there was no easy alternative. You had to go and provision your own machines to then setup.

Is it just me or ECS is relatively complicated? I mean, it is no k8, but still you need to set up a lot of stuff to get it to work.

I think it took me maybe a 1/2 day to grok the majority of it and get a build pipeline setup. Most of that time to be honest was getting the permissions correct between the build pipeline components. I used to feel the same but found it to be not that true once using it.

Github Action (Build Docker Image) > Push Image to AWS ECR for image storage > Kick off Deploy of ECS.

You can configure your ECS cluster using the console UI or something like Terraform. Once configured its set it and forget it.

Honestly not that much more complicated compared to Heroku.

Re: Is Tableau Dead?

#104

Post by prepend »

Anon_Admirer wrote: post #29

There seems to be a lot of shade here at Power BI. Forgetting Microsoft’s dominance in the market, I’ve much appreciated Microsoft’s approach to data visualization which is soft on the visuals, but gradually improving them and hard on the programming aspects, such as Power Query or “M” and Data Analysis Expressions (“DAX”). The former they have opened to anyone to create via Power BI visuals SDK. The reason I say thi…

My biggest complaint with powerbi is that there’s no versioning and collaboration across multiple authors is really hard. I can’t put it in git other than a binary blob and so that means for anything serious I have to just hope nobody goes in and breaks stuff.

At least tableau project files are just xml so you can stick it in git and have ugly merge requests. And rollback to versions if necessary.

The fact that Microsoft doesn’t even see this is a problem is infuriating and makes it hard for any serious data work where you need reproducibility and auditability.

Re: Is Tableau Dead?

#105

Post by infecto »

dkersten wrote: post #101
DanielHB wrote: post #96

Earlier quoted context omitted.

Is it just me or ECS is relatively complicated? I mean, it is no k8, but still you need to set up a lot of stuff to get it to work.

I actually gave up on AWS forever when trying to use ECS. It was incredibly complex and it kept failing to deploy, but cloudeatch only contained log entries for it about 20% of the time. All I could do was click redeploy over and over — sometimes there were logs, usually not. And they charged me for every single deploy. After a day and a $100 bill later, I gave up and used DigitalOcean’s hosted docker instead. I had…

Strange, maybe billing was different when you used it. ECS only charges for usage, you must have been requesting a lot of resources to rack up a cost that quickly. For example Ohio spot pricing is $0.01260298/hour/cpu. Deploying costs nothing.

It is a tool like any other and it is not as simple as DO's but its also a lot cheaper and more powerful once you learn the tooling. The logging on deployment is true but I think the majority/all issues I have seen with that is when the image is not configured correctly and cannot run when testing locally.

Re: Is Tableau Dead?

#106

Post by wenc »

3abiton wrote: post #26
wenc wrote: post #13

I've been using Tableau since 2013 to deploy dashboards, which in my case are front ends to ML model output. (for me, I used it as a Streamlit before Streamlit existed) People think of Tableau as a dashboard tool, which it is. But it's also a great tool for doing EDA (exploratory data analysis) on multidimensional data. Instead of writing lots of seaborn/ggplot code (and having to remember the syntax), you can just m…

But there are decent OSS alternatives that can rival it's functionality, so why go with Tableau and it's baggage?

Could you list a few? I tried looking once but nothing came close to Tableau.

Re: Is Tableau Dead?

#107

Post by wenc »

rcbdev wrote: post #32
wenc wrote: post #13

I've been using Tableau since 2013 to deploy dashboards, which in my case are front ends to ML model output. (for me, I used it as a Streamlit before Streamlit existed) People think of Tableau as a dashboard tool, which it is. But it's also a great tool for doing EDA (exploratory data analysis) on multidimensional data. Instead of writing lots of seaborn/ggplot code (and having to remember the syntax), you can just m…

Out of curiosity, how do you think Tableau compares to data mining tools like RapidMiner for EDA?

I’ve never used RapidMiner so I couldn’t say. So Tableau isn’t a full fledged statistical tool. It does have all the basic statistical functions but it doesn’t really focus on that. What it is is a visualization tool. You can actually see and touch the data in multiple dimensions by slicing and dicing it.

Statistical tests are good but they are summary statistics. Often you can get a lot of insight by just looking at the raw data. Tableau lets you see the actual outliers and ask yourself why they are there and drilling down to find out more.

The simple act of looking at the data as-is and asking the questions — before jumping to statistical testing — is an underrated discipline because it seems too simple and unsophisticated. But it yields insights that many statistical techniques will not.

Re: Is Tableau Dead?

#108

Post by data-ottawa »

c-linkage wrote: post #73
monkeydust wrote: post #54

I was an early Tableau adopter and promoter in my org many years back. At the time the incumbent, Qliksense, was lagging badly and there was genuine grass roots support for Tableau as our next gen BI tool. Adoption grew significantly and people were happy...for a while. I remember reaching out the founder/CEO at the time and getting responses back. It was great. Fast forward today and its different, over time Tableau…

Speaking of too many dashboards, my opinion at this point is that Tableau is where data goes to die. I have so many dashboards that are created in my organization but each dashboard gets one or less views per month. No one ever looks at these dashboards that get created. My feeling is that these dashboards were probably more like one-time reports that someone decided to automate. But there's definitely a sense of Tab…

Dashboards have a very long tail depending on use case.

You've got daily driver exec and performance dashboards, then you have over time trend dashboards, and finally you have drill down self serve explore of data export tools. Alerting should be separate.

It's expected that most dashboards will not be checked often, trends don't change daily so it's usually not meaningful to look at most metrics daily. Self serve tools are only needed occasionally, but exist as a cost savings on data teams doing low effort high time tasks.

Dashboard tools and spreadsheets are the only good tools for sharing data, I find most reports and spreadsheets are actually just one offs.

Re: Is Tableau Dead?

#109

Post by BandButcher »

jherskovic wrote: post #84

I was a big Tableau champion at our org. Lots of dashboards, led a user group, was the go-to expert for a while (think 10 years ago). For us, what killed it was the pricing. They got very greedy very fast. Justifying a 4-digit license, per desktop, was the real killer. We just couldn’t, so the creation of dashboards became a tool of a “priesthood” of sorts. Everyone loved the way the dashboards looked and performed,…

Haha funny hearing the other side of the story to this, i was a front end dev during the tableu peak who helped create several custom dashboards in JavaScript...for multiple different companies!

I worked on so many that eventually i came to hate making them lol, data viz can be extremely subjective and you get into weird paradigms of how much data viz you display vs user actions on that snapshot of data (ex. viewing details, exporting to diff formats, printing the dashboard with no cut-offs, manually refreshing stale data vs auto refresh)

Re: Is Tableau Dead?

#110

Post by rawgabbit »

wenc wrote: post #13

I've been using Tableau since 2013 to deploy dashboards, which in my case are front ends to ML model output. (for me, I used it as a Streamlit before Streamlit existed) People think of Tableau as a dashboard tool, which it is. But it's also a great tool for doing EDA (exploratory data analysis) on multidimensional data. Instead of writing lots of seaborn/ggplot code (and having to remember the syntax), you can just m…

>I've come to appreciate that it is beyond most people

I had the same experience. Before, my company tried several times and failed to implement enterprise dashboards and reporting. I was able to champion Tableau and get it off the ground and even the board of directors peruse the dashboards. I thought I had a huge win. I could not believe the amount of negativity I since received.

That is I believe enterprise reporting is essentially a people problem. The people who looks bad because of the metrics will always blame the tool. They won't be happy until they can manipulate the tool and massage the data until the metric becomes meaningless. In other words, it is convenient to blame the reporting if their report card shows they are failing.

Post reply on HN