Posted: Mon Mar 31, 2008 12:49 pm Post subject: Data Analysis
NeoOffice Calc needs to have a Data Analysis Toolpak similar to Excel's version. Even an updated version of the OOoStatistics offered as addon to Openoffice would work. It may need to be updated with each NeoOffice update, because the existing OOoStatistics addon will not work on the current version of NeoOffice.
Currently, users like myself have to transfer our files over to Excel in order to run these statistics. If the NeoOffice development team wants to continue to offer something equivalent to or better than MS Office, this would be considered an important feature to add.
Posted: Mon Mar 31, 2008 12:58 pm Post subject: Re: Data Analysis
Okie wrote:
NeoOffice Calc needs to have a Data Analysis Toolpak similar to Excel's version. Even an updated version of the OOoStatistics offered as addon to Openoffice would work. It may need to be updated with each NeoOffice update, because the existing OOoStatistics addon will not work on the current version of NeoOffice.
Have you contacted the developers of OOoStatistics and asked them to upgrade their code to work with the NeoOffice and OpenOffice.org 2.2.1 and higher?
Realistically, this is the best option available as the feature that you want is an advanced feature of MS Office and due to our very limited donations and developer resources, it is not within our ability to take over orphaned open source projects or to replicate every advanced feature of MS Office.
In our view, NeoOffice's purpose is not to "be equal to or better than MS Office" in every feature. Instead, our purpose is to handle the most common subset of MS Office features that we can given the resources funded by our users.
In most ways, NeoOffice is much better than MS Office. It's much more flexible and provides so many features not provided by MS Office. Overall, I have found NeoOffice to be unusually versatile.
However, my goal is to get completely away from MS Office with my personal computer. This one feature, Data Analysis Toolpak, is the only remaining item to complete the transition away from MS Office - at least in my personal life. However, I understand the problem with limited resources, so my suggestion is just that; a suggestion.
I am a new user to NeoOffice and have donated a little to the NeoOffice cause. The developers appear to stay on top of things and continually improve NeoOffice. While they continually improve NeoOffice and as I begin to regularly use more features in the software, I plan to continue to donate.
Posted: Mon Mar 31, 2008 4:52 pm Post subject: Re: Data Analysis
ovvldc wrote:
In all fairness, that particular subset has been growing steadily .
Maybe I misunderstand what you are saying, but I have only seen Okie requesting this feature. I understand that it is a critical feature for Okie's work, but since we are talking about multi-variant regression analysis, will even 30% of NeoOffice users be happy if we drop our work on the various MS Office import bugs that are in Bugzilla to learn and fix someone else's broken and abondoned code provides a feature that most people (including myself) don't use or understand the underlying math? Since we also already have single variant regression analysis support, I really have a hard time accepting that even 1% of the NeoOffice user base needs this feature on a day to day basis.
My point was a nice way of saying that we don't have the resources to solve all of the world's problems and when it seems likely that only a very tiny portion of our users would benefit (or even notice) a particular feature, it seems very unfair to ignore the known issues that a vast majority of our users are suffering with and use their donations to implement a feature that they don't want or need.
Joined: Jun 20, 2006 Posts: 2051 Location: Midwest, USA
Posted: Mon Mar 31, 2008 5:11 pm Post subject:
I think the subset Oscar was referring to was the subset of MS Office features Neo can provide, and commenting that you've added a series of such features lately. (Menu improvements, grammar checker access, etc.) But maybe I've misread it?
I think the subset Oscar was referring to was the subset of MS Office features Neo can provide, and commenting that you've added a series of such features lately. (Menu improvements, grammar checker access, etc.) But maybe I've misread it?
Yeah, I thought that too. But I thought I'd cover the flip side as I don't really want Okie to feel that I am dismissing the feature request out of hand. I know, it is probably overkill, but I believe that the developer of the OOoStatistics extension is probably the best person(s) to provide this feature given their expertise and previous work.
But just in case I am misunderstanding what is wrong in OOoStatistics, I can install it in both NeoOffice and OOo 2.2.1, 2.3.1 and 2.4 and see if it only fails with NeoOffice. If it only fails in NeoOffice, then it is our bug and we'll deal with that.
Okie,
Can you post a list of steps that I can use to use the macro and get the same error that you see?
No, that is what I meant. And I agree that multivariant statistics is not for everyone. Though SPSS is one of the most horrible pieces of software I have ever had to work with.
-Oz _________________ "What do you think of Western Civilization?"
"I think it would be a good idea!"
- Mohandas Karamchand Gandhi
I don't know why the Data Analysis package is considered an advanced feature. I know the Multiple Regression part of the package was available in the original Lotus. The first version of Quattro Pro included it as well. Even the earliest versions of Excel had Multiple Regression available. I have been using Multiple Regression function on spreadsheets for nearly 18 years. It was commonly available on nearly all popular spreadsheet software. However, it now seems to be only available on Excel and Quattro Pro.
Yes, that's probably about right that only about 1% of the users would regularly use the Data Analysis package. However, there are a number of other features already available in NeoOffice that only about 1% of the users regularly use. The Data Analysis package is simply a nice feature that I have been accustomed to having available while working on spreadsheets. I use it frequently at work and at home.
Anyway, below are general instructions for installing OOoStatistics and running Multiple Regression. This is the one I am most familiar with and most important for my purposes.
After downloading the OOoStatistics installation file, open it with NeoOffice.
Install OOoStatistics according to the developer's instructions.
Close NeoOffice completely, and open it back up.
Open New Spreadsheet
Open OOoStatistics installation file.
Select Multiple Regression worksheet
Copy columns A through D and paste to New Spreadsheet
In New Spreadsheet block all data. Note: In this example there is text in the top row. It automatically recognizes these as labels. If there are only numbers in the top row, it will check to see if they are supposed to be labels. However, it will not give you a choice but will the top row as labels. That is the first glitch, but there's an easy workaround.
Select OOStat, Basic Stats, Multiple Regression
Next glitch: Note than in the output, it will not allow a Range to be inserted. When I run it, it will only allow output to a new sheet.
Select New sheet and keep the default name. I generally prefer using the Range by placing the output off to the side of the data, but this option of course is not available for me.
Select Okay
In the next window that pops up, select “Fitâ€
Last glitch: The window with the macro code will pop up highlighting where the error occurred. Then an error window will pop up explaining the error. Somehow the calculation in the code is dividing by zero.
Thank you for the steps. Sorry to give you the bad news, but the problem is not NeoOffice. It is a plain old bug in the OOoStatistics macro code as the same error happens in OpenOffice.org version 2.2.1 through the 2.4 "DEV300_m2" build.
Okie wrote:
Yes, that's probably about right that only about 1% of the users would regularly use the Data Analysis package. However, there are a number of other features already available in NeoOffice that only about 1% of the users regularly use. The Data Analysis package is simply a nice feature that I have been accustomed to having available while working on spreadsheets. I use it frequently at work and at home.
OK. Let me explain the economics. Most of the features in OpenOffice.org were funded by a company with over $10 billion in annual revenues and that company has several billiion in cash reserves. They are the ones that paid for the features that less than 1% of users use.
In contrast, NeoOffice exists on donations that only support myself. I work 7 days a week for 12 hours each day doing coding and customer service and Ed works nights and weekends for no pay. To be blunt, my paycheck depends on donations continually coming in which is why you see NeoOffice is focussed on general features on not on features that benefit less than 1% of the users.
In monetary terms, it comes down to this: the cost to learn the math behind all these functions, then learn and fix and support the broken code that someone wrote and abondoned is not likely to ever be recouped through donations. This means that the cost to me and Ed will likely be many many times higher than the cost of a $150 MS Office 2008 license is to you and the people like yourself who need a solid statistical package.
Given the learning curve involved, "little" features like this will probably require no less than US$25K and maybe more. I don't have that kind of money sitting around burning a hole in my pocket, but if you do and you think this is critical for NeoOffice and you have the funds to make this happen, by all means contact me.
Even if you have that kind of funding, I would still recommend that you talk to the developer who wrote OOoStatistics. They know the code and the math so fixing and supporting their existing spreadsheet macros would likely be a mere fraction of what it would cost us and we can discuss a grant agreement for this feature.
I appreciate the time you took to look at this. I completely understand the issues and the reasons. I simply ask that when the resources become available that this could be considered with some future upgrade.
In the meantime, I plan to play around with the macros in the OOoStatistics to see if I can correct the problem. I understand the math involved, but not all the code. It will be a fun little project to see if I figure this out.
If you do figure out a way to fix this, we would be very happy to hear it. And I am sure we could help find a place to host the updated statistics package, so that future users will not have to go through the same problem..
Best wishes,
Oscar _________________ "What do you think of Western Civilization?"
"I think it would be a good idea!"
- Mohandas Karamchand Gandhi
In the meantime, I plan to play around with the macros in the OOoStatistics to see if I can correct the problem. I understand the math involved, but not all the code. It will be a fun little project to see if I figure this out.
If you know the math, you probably know the more important of the two to know.
I will let you know if I do figure it out. A few years ago I worked with macros a lot. Whether or not I'm successful with this, it will definitely be a good refresher.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You cannot download files in this forum
.
