Forum Discussion
Hide/Show Column in Table Visual Matrix on Report
Everyone needs to make their own call if a product is right for them. But surely if there are too many issues for you to count, you should be able to mention, say just 2 more in addition to this thread. The one you posted above is not a work around. The default sort of every column is alpha numeric. There is a clearly defined and robust process to change the sort order - you simply specify the order you want, and then it will use that sort order. I'm not sure how else it could work. Power BU cannot just "know" the sort order you want.
I'm just trying to understand what it is that makes the tool unworkable for you.
I've been out of the loop for about two years, and since then a lot of water flowed under a lot of bridges, so I think I should be to be excused for not being able to come up with more examples.
But just to focus on the problem in that other thread: I always believed the product has to serve the user, not the other way around; so while I appreciate the fact that the "default sort of every column is alpha numeric" - the default is a problem for me as a user. So the product needs to either change the default to "not sorted", or offer a simple check-the-box option of "no sort". Please take look at Message 15 of that other thread and tell me if you sincerely believe that what is described there consititues "a clearly defined and robust process to change the sort order".
It's strange to me that, after all the time that has passed, I still feel greatly annoyed when I read that exchange...
So you stopped using Power BI 2 years ago - that was not clear in your post yesterday. It has progressed since then, for sure.
Regarding the sort order "issue", Power BI is a database. Databases (all databases) have no concept of row order. All data is loaded and stored in a table, then retrieved when needed. The order the data is stored and retrieved is not important and is not tracked. In fact Power BI stores the records in order to acheive high compression rates. That is why 100MB of source data can often be loaded into a Power BI database that is around 10MB (10x compression is common). If order of the data is important, then you need to load a key that indicates the order you want. That is how the sort order is managed in all databases including PBI.
I never said that a question/answer thread in a public forum was "the official" clearly defined process. Here is the official documentation. https://docs.microsoft.com/en-us/power-bi/create-reports/desktop-sort-by-column
It is also strange to me that you feel so annoyed by this given there is a simple, scalable, standardised solution.
I have watched this product develop from zero to where it is today over the last 5 years. Personally I did not use it for 2 years as I felt it was not mature enough. It was around October 2017 that I felt it was good enough to at least consider. It has had 150% additional development effort since then.
I'm not saying this product is right for you - I have no way of judging that. But you did just post to say that you walked away because of countless issues without qualifying that your assessment was 2 years old, and without providing details as to what those countless issues were/are. If there are still such "this product is unusable" issues, I would like to know what they are. I lend my weight to improving the product all the time, and if there are gaps that do not have focus, I'm more than willing to help. But if there are not countless issues that make the product unusable, I think people reading this threads should know that too.I don't think we're going to solve these problesm tonight/today, so allow mr to make one narrow point (to get it off my chest) and then go more general.
Specifically, about the link you provided for the official clearly defined sorting process is indeed what you say it is. But even looking at it from the vanatage point of someone has been absent from the scene for a long while, it's clear it doesn't address the main (to me) issue: how to prevent any type of sorting. While I understand the need of PBI or other databases to store, compress, slice and dice data - on occassion (not frequently maybe, but still) data is immutable, sacrosanct, etched in stone. The system should be able to do all those things to the data for its own convenience, but it should also be able to output the data exactly as it was input - no surgery allowed. I am not a DB designer and have no idea how to do that, but maybe at the time data is loaded - there should be some way of indicating a "do not touch" intent. If you could lend your weight to making that possible, that would (in my opinion) be an important contribution to the cause.
On the more general level - you may actually be right; it's entirely possible I've been away long enough and significant improvements may have been made to warrant a re-examination on my part. As I said, I was positivley disposed toward PBI and really, really wanted it to work for me. So what I would probably do is take some time, reinstall, try to do some of those things I need to do frequently and see if this time things go a little more smoothly.
In the interim, Matt, thanks for your time (and attention!). It was really informative.