Comments
2 comments
-
Hi there.
I'll try and take as many of your points as I can.
There are a few reasons we went down the route of the grouping bars in the top results grid. When we met with people and when we tested the product, we observed that the first thing most users would do (in version 3 or 4) would be to group via the difference type. Then they'd follow this by getting rid of the identical objects so that they only had the difference types that they were interested in displayed.
The functionality of being able to hide all of a difference type, can inadvertently cause other problems, specifically when filtering. We found that new users found it very hard to know whether a filter would exclude the object from synchronization or if it would just hide it from view.
We wanted to make projects more useful, such that they would remember a lot more of their settings between sessions. The side effect of this is that if you had excluded a difference type, and subsequently a difference of this type was detected, the user wouldn't know about it. Keeping the grouping bars there, means that you can never inadvertently miss a difference that you weren't expecting. This is highly useful when a lot of our users run through SQL Compare pretty much on autopilot. The grouping bars therefore allow you to get a good overview of your differences without any scrolling - which then informs your decision over what to investigate.
So we felt that it was much more important that you'd always be able to know the extent of your differences without the risk that a difference would go unnoticed because of how you had left the project previously.
Having said all of that, we'll certainly look to include the ability to not have the grouping bars shown, but this won't make it into the initial release of version 5.
Regarding filters, this is a route we've been forced to take because of the huge increase in the number of object types that SQL Server 2005 has introduced. Having these along a toolbar would mean that we'd be forced to show the icon only, and when you have so many, finding the specific one you want becomes a painful tool tip search. It's far quicker to be able to read the object type in a text list and get a clear indication whether it is on or off - even if you've had to open a panel to achieve this.
The headers you mention in the difference pane, e.g. extended properties, columns etc. have always been there, but they were very easy to overlook and so their functionality was lost. They are useful though, in much the same way as the main grouping bars in the top pane, so we just made them more visible.
We've changed the behaviour of the copy actions button.
In all the performance tests that we've conducted, we've been faster. If there are specific instances where you know of a competitor's product outperforming us, we'd be interested in being able to replicate it.
Regarding Data Compare:
We've worked hard to ensure that the product (like SQL Compare) has a strong and consistent use of symmetry. In Data Compare, the middle three columns are there to help give a spacial feel as to where the data resides. If it's in the middle column, it's in both, but it differs, if it's on the left, then it's in the left hand database only, same goes for the right. This does mean that with the combination of the interactive help, tool tips and icons, we've been able to see user's understand this concept very quickly. It's clearly different from version 4, but it's much easier for new users to get to grips with.
Application options have been standardised between Compare and Data Compare.
You can clone projects that you have previously set up with specific options, so once you have a set of options you like, clone a project instead of creating a new one and you'll keep all of your options.
I hope that in part has explained some of our reasoning and thank you for your comments - you've made a lot of valid points. I'll make sure they get fed back into development.
Dom -
Thank you for the response. I really appreciate it, and as I said I was trying to be constructive.
You are correct that most users (including myself) usually hide equal objects. I do not object to them being there. In fact in Data Compare, I really like that you show even objects that were not compared due to a missing key. That is such a huge improvement.
However, It would be nice to be able to remove the grouping up above - at least of the 3 different types (missing, additional, different) - even if equal remained a separate group. Or maybe if the headers could be made thinner - they just are way too bulky if you have only one or two objects in the difference list. They aren't as clumsy if you have large amounts of differences in a section, but if you are constantly keeping databases synch'd you most likely have only a few differences and then the headers overtake the actual data.
It seems it should not be too hard since you have that ability for the bottom section (the script)
I do understand the more objects and the purpose of the sidebar - I see it being much more useful for new users as well. However, I think it would be a nice option to allow advanced users to choose to display them as a thin toolbar across the top. That way once you are used to the icons (especially since they are visible next to the objects being compared) you can choose to have the quick toolbar, but when you want a detailed list you could still choose to have the side panel.
It could be a simple option like "Show filters toolbar" in the option menu.
Just something to consider.
The idea of closing a project is very nice. I also like that the projects auto-save when you close the program.
>>We've changed the behaviour of the copy actions button.
What does that mean? DOes it mean you added the ability to add the "Copy SQL Script"?
It is interesting that you feel the new look in Data Compare is easier for new users to "get grips with". I am a relatively new user (have only been using your products since last Fall). I was also the one to evaluate products to determine which one to use for our office. I found the old interface so intuitive over competitors. I am having a very hard time with this new one. So, I also showed to another programmer in our office. Overall he was very impressed with the new interface, except this portion. He found it very hard to tell where there were differences and I had to point them out.
I also ran a test where we specifically introduced a difference and tried to find it. It was VERY hard to do so - for either of us - even knowing where the difference was.
Also the bottom section that shows the data difference is very hard to see - even knowing where the difference is.
Overall I think that there are many nice improvements, a few things that in practical use become clumsy, and mainly one thing (specifically the Data Compare comparisons) that is much worse. However, I think those could be easily fixed.
Add comment
Please sign in to leave a comment.
==========SQL Compare================
I think there are some improvements:
1) The main toolbar across top for editing project, refresh, etc
2) The very clear visual indicator of the databases being compared and current synch direction
However I think the version 4 interface was significatnly nicer in the following respects:
1) The comparison section was MUCH easier to work with. It was nice to have all objects listed together and then be able to turn on or off which types of comparisons you were looking at (i.e. equal, missing, different, additional).
It is very bulky now and MUCH harder to tell what the difference is.
I do not mind the greying out of equal objects. But that could be in the same section. It is very hard to tell with objects with differences. The header sections take up way too much room.
Perhaps a good option would be for you to allow users to pick between 2 interfaces. One with the bulky headers, and one more like the previous version where all items were listed, but a nice clear icon was used to indicate what the differences were
2) One of the nicest things going from previous versions to 4 was the toolbar across the top that contained the filters. Now it is a very bulky side panel that takes up a lot of screen space. Additionally it is not always there (and can't be with how bulky it is), but the nice compact toolbar previously was very nice.
3) While I think it is nice that you are trying to group changes with friendly headers (such as "Extended Properties", "Columns", etc), I think the headers for these are way too bulky. They take up too much vertical space. I think they would work better thinner.
I notices that in this section you do have the option to view or not view the header groups. I think this is a definite plus and should be available for the top section as well.
However, I did notice that it did not remember the settings. When I unchecked the "show in groups" for one project (since you can't set before that point that I could see), then created another project, it did not remember the setting. However, it did remeber the setting within a project
4) When actually performing a synch at step "3. Summary", there is a button called "Copy Actions". This is not so useful to me, but I don't oppose it being there. However, I think that it would be VERY helpful to have a "Copy SQL Script" button in addition to the "View SQL Script" button, which automatically copied the script to the clipboard.
5) BUG: A bug I notices, which was also present in version 4 that I currently have a thread going on about: It incorrectly sets up to use Query Analyzer (SQL 2000) when only SQL Management Studio (SQL 2005) tools are installed.
I do like the ability to change this on the fly
6) I have seen comments saying the comparison speed was improved. I have not noticed that. It seems slower for all of the comparisons I have done. Not significantly slow, but noticably so. It used to out-perform competitiors, but seems about the same speed now.
The actual synch however, still seems as fast as before.
The refresh after a synch does seem to be faster.
7) It might be nice to have an X in the right corner to close the current project instead of having to use the File Menu to do so.
8) One thing that has never been done and is a suggestion if possible, is when synchronizing a user - I understand that password cannot be synched, but can the default database be synched? I see no harm in that and more harm in how it currently is - which defaults the users to the master database.
=======SQL Data Compare===========
Comment#7 above applies to Data Compare as well
1) I also feel the header sections here are bulky. SImilar comments to above.
2) The icons for "The number of records which exist in ____" is not very clear. Nor is the mouseover very helpful since it only includes the database name, and often people are synching datab between a development and production server so the database names will be the same in each, while the server names will vary. Perhaps including the server name would help. Or the blue/red directional arrows were more indicative of the changes than the new icons.
3) The application Options seem to be in a different order than in SQL compare. They should be consisntent across your applications where possible.
4) It would be nice to be able to set the project configuration options once and have them apply as the defaults for all new projects. Currently if you have custom settings (such as do not ignore whitespace differences) you have to set up for each project every time you create a new project.
5) When I had two extra records in the table that was being synch'd. I unchecked on record. Then I went to the synch wizard. It listed in the action plan "Delete 2 rows" It should only have said Delete 1 row since only 1 was checked. However only 1 was actually deleted. So the synch worked properly, but the action plan and the modification screens displayed incorrect info.
6) It is a LOT LOT harder to see differences between records. I see a few reasons for this:
a) The break apart of each column into a isolated group disturbs the flow reading across an entire record
b) The highlighting of the field is so similar to the background color. The bright blue and red previously was much easier to discern.
c) The fact that one table is left justified and the other is right justified makes it much harder to compare longer values
7) I tis strange that in SQL compare identical records are greyed out and in SQL Data Compare they are still black.
Personally I prefer black, but don't mind the grey out terribly. However I DO mind the inconsistency across products that are part of a bundle.
I hope you take these comments as constructive criticism and can incorporate some of the changes into the version 5 before its actual release. Please ask if you would like me to expand on any of these, provide details, screenshots, etc..