TechCrunch posted one interesting article Kevin Merritt wrote (CEO and founder of blist), on how blist was used in Obama-Biden Transition, to disclose the names of donors to the transition project. Congratulations to Kevin as his product was chosen by Obama team over the competing services.
What I want to clear up is how other web-based products would handle this, in case the choice was made in favor of products I review in my blog.
So I exported data from Obama-Biden-Donor-List and tried to test following services capibility to handle it.
I managed to import the data in 6 copy/paste shots, importing 10 000 records each time. This is how it looks:
I have such an impression that Dabble DB works a way slower with such data volume. At least as I tried to navigate within the table. Unfortunately Dabble DB doesn’t have simple ways for Obama Team to insert the list in their blog. To implement this they should have used Dabble DB JavaScript API.
There was no problem importing data in one step. I used the file I’ve got exporting data from blist. This is the result:
In TeamDesk I don’t see any simple way to display the list of people who donated thorough their site as well. The developer offers to use API for this.
No luck with Zoho Creator. There was no way to check whether Obama Team could use Zoho Creator as the system couldn’t parse the app correctly through import file:
But another Zoho product - DB & Reports gives the opportunity to do that. This is what I’ve got:
This product also has built in tools for publishing the data through your blog or web site.
Its Coghead turn now to prove its value. This is how the result of data import looks:
To integrate results in your own site one can use Coglets.
If Obama Team choice was TrackVia the result of data importing would look like this:
Using "Publish this view" feature one can display it on his site.
If use LongJump this is the look of data within the system:
As I haven’t found in LongJump simple means to display the list on my site, I suppose we should use API within this program as well.
Unfortunately I couldn’t check how to implement this in QuickBase for my trial has already expired.
Anyway, I think each vendor won’t mind to be chosen by Obama Team and definitely each service provider would improve the system and make needed changes to satisfy their requirements.
Nobody knows what triggered the decision in favor of blist or it was just a random choice of the system, but that must be huge for the vendor.
I can hardly imagine any better promotion for the product.
Testing of web based database software in solving real-life business tasks and the best ways of its implementation. Finding specific solutions for specific problems.
Tuesday, January 27, 2009
Friday, January 23, 2009
SaaS lock in is a real lock in!
The silent agreement of vendors not to mention the fact once you get into the service and this is it, you are locked in, works pretty well. Who wants to scare off potential users? Moreover, some of them claim it’s feasible to transfer the functionality and there is nothing to worry about. Well, there is a reason.
Folks are not that eager to entrust data and logic to SaaS vendors to be locked in.
On the one hand, vendors don’t want to extend their service, aiming to solve this issue at the same time to make their product the most palatable for the customers.
Take into account, developing the app within any system there is no way the functionality you have tuned can be transferred to another one. Its inevitable one day the user finds he is locked within the particular system.
And another striking demonstration of that is PC Magazine award. Having chosen the system in 2005 they forced to be happy customers by now. Isn’t it a lock in?
There won’t be a big catastrophe if traditional software developer just stops providing its support, the old version will be running anyway and the user have enough time to replace it. But in case SaaS vendor informs you in a month the service will be down the situation is a way more dramatic.
This is exactly why the choice of vendor is that important. So, you’d better think twice. If take a chance, make sure it’s really worth it.
Folks are not that eager to entrust data and logic to SaaS vendors to be locked in.
On the one hand, vendors don’t want to extend their service, aiming to solve this issue at the same time to make their product the most palatable for the customers.
Take into account, developing the app within any system there is no way the functionality you have tuned can be transferred to another one. Its inevitable one day the user finds he is locked within the particular system.
And another striking demonstration of that is PC Magazine award. Having chosen the system in 2005 they forced to be happy customers by now. Isn’t it a lock in?
There won’t be a big catastrophe if traditional software developer just stops providing its support, the old version will be running anyway and the user have enough time to replace it. But in case SaaS vendor informs you in a month the service will be down the situation is a way more dramatic.
This is exactly why the choice of vendor is that important. So, you’d better think twice. If take a chance, make sure it’s really worth it.
Thursday, January 22, 2009
Blist turnes to own method for editing the same record by multiple users. I had no idea!
Record editing by multiple users is handled absolutely differently in Blist. I think it’s really worth mentioning and taking a look at. This is the web list I've chosen to check multiple users editing the same record.
And this is how the issue is solved. The approach is different. Any data modification by any user is sent at once to the server and all users can take a look on changes the data is subjected to right now. Updated data appears on their screens. This is how it works in real life.
As one can see form screencast it’s difficult to define if the method works well in case we have the Form and Table. The very approach is really impressive, but I would question the usability for common users (I mean taking control of data modifications) anyway.
And this is how the issue is solved. The approach is different. Any data modification by any user is sent at once to the server and all users can take a look on changes the data is subjected to right now. Updated data appears on their screens. This is how it works in real life.
As one can see form screencast it’s difficult to define if the method works well in case we have the Form and Table. The very approach is really impressive, but I would question the usability for common users (I mean taking control of data modifications) anyway.
Wednesday, January 21, 2009
Slow down while editing the same records by multiple users in LongJump
This record editing can be a tricky thing! And it really is. Just checked how multiple users editing works in LongJump. It didn’t take a lot of time. I went through few things.
Opened one and the same record for editing in IE and Firefox. In FF I changed First Name and Birthday:
And in IE I changed First and Last Name:
So, let’s save these changes in IE first:
And now in Firefox:
Result:
So, all changes were saved without any warnings or questions. Not that good news for the common user, who probably will blankly stare at the screen wondering what the heck is going on? Yes, LongJump doesn’t have any built-in functionality for simultaneous editing. Be advised.
Opened one and the same record for editing in IE and Firefox. In FF I changed First Name and Birthday:
And in IE I changed First and Last Name:
So, let’s save these changes in IE first:
And now in Firefox:
Result:
So, all changes were saved without any warnings or questions. Not that good news for the common user, who probably will blankly stare at the screen wondering what the heck is going on? Yes, LongJump doesn’t have any built-in functionality for simultaneous editing. Be advised.
Tuesday, January 20, 2009
Mission Impossible: Zoho Creator find and merge duplicates
Good structured data is a crucial factor of efficiency and success while working within any web-based database system. As the practice shows, unfortunately, the data is affected oftentimes when many users work within the application.
Really, it’s quite a task to get rid of data bugs when the system is used by many users simultaneously.
But what makes any system to be considered as good one is its tools to fix it. This is exactly why I think find and merge duplicate records function is that important.
About a month ago I tried to find out how this function is implemented in Zoho Creator, with no success. I expected to get any answer form Zoho support, but I guess there is no way you can do it, this is the only point I see why Zoho support guys are still silent.
Having analyzed Zoho blogs and forums I can say Zoho CRM is the only product that supports Find and Merge duplicates right now.
I don’t understand why some developers prefer simply not to answer the request, but stay silent or dodge the question if the system lacks some functionality.
Really, it’s quite a task to get rid of data bugs when the system is used by many users simultaneously.
But what makes any system to be considered as good one is its tools to fix it. This is exactly why I think find and merge duplicate records function is that important.
About a month ago I tried to find out how this function is implemented in Zoho Creator, with no success. I expected to get any answer form Zoho support, but I guess there is no way you can do it, this is the only point I see why Zoho support guys are still silent.
Having analyzed Zoho blogs and forums I can say Zoho CRM is the only product that supports Find and Merge duplicates right now.
I don’t understand why some developers prefer simply not to answer the request, but stay silent or dodge the question if the system lacks some functionality.
Friday, January 16, 2009
Giving Blist a chance
I have considered Blist as simple single/flat list oriented system. But skimming their web site once again I discovered that developers offer to use it as a solution for organization of CRM and Project Management.
So I decided to practically check out what one can do within the system and started with large volume handling.
It turned out through Blist one can create his list directly through import from Excel file, so let’s check how it works.
The system correctly recognized the data structure, but the column types were not detected correctly(all columns have the Text type by default):
I manually changed column types and continued to import the data.
The import process is conducted in a background mode, so meanwhile the user can go on operating within the system. Well, it’s already something.
This is the look of import results:
Unfortunately, while data import Blist had some failure, despite the fact there were no notifications about errors, no warnings that something goes wrong. As from Fax column the data were shifted and partly disappeared even though everything was reflected on the preview correctly.
In general I’d say working with of such volume as 20K I had no navigation or data reflection problems.
Conclusions:
It’s a great advantage Blist is capable of direct Excel file importing and asynchronous data download. There was no problem with handling of such data volume as well. I hope the developers can cast light upon the failure I experienced importing the file. Anyway, the service is on the beta stage right now, and I believe my tests can only be of a help in system enhancement.
So I decided to practically check out what one can do within the system and started with large volume handling.
It turned out through Blist one can create his list directly through import from Excel file, so let’s check how it works.
The system correctly recognized the data structure, but the column types were not detected correctly(all columns have the Text type by default):
I manually changed column types and continued to import the data.
The import process is conducted in a background mode, so meanwhile the user can go on operating within the system. Well, it’s already something.
This is the look of import results:
Unfortunately, while data import Blist had some failure, despite the fact there were no notifications about errors, no warnings that something goes wrong. As from Fax column the data were shifted and partly disappeared even though everything was reflected on the preview correctly.
In general I’d say working with of such volume as 20K I had no navigation or data reflection problems.
Conclusions:
It’s a great advantage Blist is capable of direct Excel file importing and asynchronous data download. There was no problem with handling of such data volume as well. I hope the developers can cast light upon the failure I experienced importing the file. Anyway, the service is on the beta stage right now, and I believe my tests can only be of a help in system enhancement.
Wednesday, January 14, 2009
PC Mag editor has NO choice. So, QuickBase awarded!
I just stumbled upon an interesting twitt that QuickBase got PC Magazine's Editor's Choice award.
Having read the article I came up with an idea the award should have been named: PC Magazine's Editor's HAS NO Choice.
Why?
Simply because in 2008 PC Magazine had not reviewed a single database system, so the very concept of choice is missing, as we can see it here.
The second argument is adduced by Edward Mendelson, in the end of his review:
Personally I consider QuickBase to be one of the best products I’ve ever tested, but this particular example of PC Mag selection has nothing to do with actual choice.
Having read the article I came up with an idea the award should have been named: PC Magazine's Editor's HAS NO Choice.
Why?
Simply because in 2008 PC Magazine had not reviewed a single database system, so the very concept of choice is missing, as we can see it here.
The second argument is adduced by Edward Mendelson, in the end of his review:
"… if you use your database as a storage area for documents. You can't easily transfer files to any other storage system. You're effectively locked into QuickBase until the end of time."
Personally I consider QuickBase to be one of the best products I’ve ever tested, but this particular example of PC Mag selection has nothing to do with actual choice.
Tuesday, January 13, 2009
What's the user licensing verge of TrackVia?
I keep on clearing up peculiarities of how web-based programs can serve best to common users and what major vendors offer.
So, lets check how TrackVia handles unlimited users issue? I mean if adding new users smatrly principle works.
Check out the pricing in the first place. Each user with the permission to access the database is considered as Licensed User.
If you need to grant access for many users in TrackVia two things will be of help here:
Website Forms – this feature allows you to create public form that adds data to your database.
TrackVia allows generating HTML for a web form, and you can copy and paste it into your web page using your favorite HTML editor. This is how the results will look like:
If you want to quit here, you can – you already have a fully-functioning public TrackVia form that adds all new records in your database. You can even improve your form’s appearance and functionality if you are familiar with HTML script, and here you can specify how.
Published views – this feature allows you to make views available to the general public: people who are not TrackVia users and who aren't logged in to TrackVia. You can do this by publishing the view into an existing web page, or by linking to or sharing a public URL. This is how it looks:
Published views do not allows modifying or viewing particular record data, but they do have working sort and pagination (results per page) links and will always show up-to-date information from your TrackVia database.
Conclusions:
TrackVia gives you good opportunities to organize public access to your database. The thing is your needs just have to fit the scope of viewing and adding the records to the database.
So, lets check how TrackVia handles unlimited users issue? I mean if adding new users smatrly principle works.
Check out the pricing in the first place. Each user with the permission to access the database is considered as Licensed User.
If you need to grant access for many users in TrackVia two things will be of help here:
Website Forms – this feature allows you to create public form that adds data to your database.
TrackVia allows generating HTML for a web form, and you can copy and paste it into your web page using your favorite HTML editor. This is how the results will look like:
If you want to quit here, you can – you already have a fully-functioning public TrackVia form that adds all new records in your database. You can even improve your form’s appearance and functionality if you are familiar with HTML script, and here you can specify how.
Published views – this feature allows you to make views available to the general public: people who are not TrackVia users and who aren't logged in to TrackVia. You can do this by publishing the view into an existing web page, or by linking to or sharing a public URL. This is how it looks:
Published views do not allows modifying or viewing particular record data, but they do have working sort and pagination (results per page) links and will always show up-to-date information from your TrackVia database.
Conclusions:
TrackVia gives you good opportunities to organize public access to your database. The thing is your needs just have to fit the scope of viewing and adding the records to the database.
Monday, January 12, 2009
Can LongJump make a big jump to beat the competitors?
Opening the year I made up my mind to try a new system – LongJump. And, as always, the first step of testing is to try the program capability of handling large data volume so one can see what to expect in further app development.
LongJump allows importing data at once, creating the object. So let’s use this option.
The system recognized columns correctly, but didn’t manage to select needed data type automatically. I had to set up that manually.
Unfortunately, the import process was not success at first too.
Though the system was unable to import the data from the first shot, the object field structure was partially created (without e-mail column…).
Anyway, I gave LongJump the second chance and used Import Data option.
Here it is. The system didn’t manage to relate fields by title according to file data, so I had to create mapping manually.
When you take the last step the system offers to set up additional options and save import scenario to speedup import next time.
The very import process is conducted in background mode so one can work within LongJump at the same time.
As soon as the system it through with import the user receives the e-mail notification of its status.
Since we have already imported the data it’s high time to check it. I must say the look of automatically generated view and value of “Modified By” column puzzled me a bit.
Let’s press Details link and take a look at our data within the system:
One can see everything is ok but Birthday and Hireday fields of the system for its empty, despite the system didn’t report on any errors and with no warning about import results as well.
After “All records” set up I checked LongJump capability to handle 20K records.
Good news: LongJump handled 20K records with no problem. I think the system is capable of handling even larger volume.
Conclusions:
On the one hand, LongJump has such powerful mechanisms as background import/export, what proves that creators developed the system to handle large data volumes. On the other hand, the error diagnostics should be definitely improved. I mean lost e-mail column, not imported data of birthday and hire date fields without any notification of occurred errors can be of an unpleasant surprise for LongJump common users.
LongJump allows importing data at once, creating the object. So let’s use this option.
The system recognized columns correctly, but didn’t manage to select needed data type automatically. I had to set up that manually.
Unfortunately, the import process was not success at first too.
Though the system was unable to import the data from the first shot, the object field structure was partially created (without e-mail column…).
Anyway, I gave LongJump the second chance and used Import Data option.
Here it is. The system didn’t manage to relate fields by title according to file data, so I had to create mapping manually.
When you take the last step the system offers to set up additional options and save import scenario to speedup import next time.
The very import process is conducted in background mode so one can work within LongJump at the same time.
As soon as the system it through with import the user receives the e-mail notification of its status.
Since we have already imported the data it’s high time to check it. I must say the look of automatically generated view and value of “Modified By” column puzzled me a bit.
Let’s press Details link and take a look at our data within the system:
One can see everything is ok but Birthday and Hireday fields of the system for its empty, despite the system didn’t report on any errors and with no warning about import results as well.
After “All records” set up I checked LongJump capability to handle 20K records.
Good news: LongJump handled 20K records with no problem. I think the system is capable of handling even larger volume.
Conclusions:
On the one hand, LongJump has such powerful mechanisms as background import/export, what proves that creators developed the system to handle large data volumes. On the other hand, the error diagnostics should be definitely improved. I mean lost e-mail column, not imported data of birthday and hire date fields without any notification of occurred errors can be of an unpleasant surprise for LongJump common users.
Subscribe to:
Posts (Atom)