The Excel Syndrome



  • When I was hired by my current employer, they had no developers on staff.  That said, whenever someone here encountered a problem in the past, the answer was almost always "make a spreadsheet".

    Need to track book check-outs/ins at the company library? Make a shared spreadsheet on the file server.

    Keep track of freelancers assigned to various projects? Make a shared spreadsheet on the file server.

    When the problem of project tracking came up, this was their solution:  Create a printed spreadsheet that is affixed to the physical copy of the project, which people write on as the project progresses.  Then create an array of spreadsheets distributed across the file server, in different project directories, each with whatever organization and formatting the project leader decides to use, copied manually into a master tracking spreadsheet which would then be printed out and hand-keyed into a 20 year old MS Access database to print out a master project schedule.  The process of keying in and creating this schedule is done every month and takes one to two weeks.

    Recently, projects have been falling behind, and people aren't sure where things are, what's late, what's on time, etc.  And the ancient MS Access database reports are transposing columns, making reports inaccurate (and impossible to debug, given the several hundred old reports that are no longer used, and the thousands of fragments of SQL that link to other fragments that link to other fragments to create the one report in the entire database that we do actually use).

    The solution, of course, is not to rethink the whole tracking methodology, but to add another layer of tracking spreadsheets that record more detailed tracking data.

    The basic progression flows like this:

    1. Problem
    2. Solution:  Spreadsheet
    3. Another problem
    4. Solution:  Bigger spreadsheet.
    5. GOTO 1

    The above steps have been iterated so many times that we now have a tracking spreadsheet that gets printed out every week (taking a couple hours to manullay fill in and format) and is 1 meter tall and a couple meters wide, which people write on by hand while still doing the process described above, manually filling out duplicate data across multiple differently formatted spreadsheets.

    If I weren't working on a system to get rid of all these spreadsheets, I can imagine that, in a few years time, we'd have a spreadsheet that takes up the entire length of the conference room wall.

    Given enough time, these spreadsheets would only be limited by the roll size of our plotter printer (which I think is somewhere around 30 meters).


  • Considered Harmful

    Add in a few screen captures of massive spreadsheets and I think we've got a winner.



  • @joe.edwards said:

    Add in a few screen captures of massive spreadsheets and I think we've got a winner.

    This definitely belongs as a front-page article.

     



  • @joe.edwards said:

    Add in a few screen captures of massive spreadsheets and I think we've got a winner.

    Here is a screenshot of the Master Spreadsheet (actually, it is several screenshots stitched together).  Only the full width of the spreadsheet has been captured (i.e. all the columns have been captured, but not all the rows).  It is in PNG format - so those blocks you see are not compression artifacts, they are text zoomed way out.  The forum prevents the entire image from being displayed; for me, it's showing about half of the spreadsheet's columns.

    [URL=http://img233.imageshack.us/i/tdwtfthemasterspreadshe.png/][IMG]http://img233.imageshack.us/img233/3119/tdwtfthemasterspreadshe.png[/IMG][/URL]



  • I now also love the Kraken.


  • ♿ (Parody)

    @KrakenLover said:

    Here is a screenshot of the Master Spreadsheet

    This reminds me of the way you tend to see DNA displayed. I wonder what sort of life form it really is?



  •  @boomzilla said:

    @KrakenLover said:
    Here is a screenshot of the Master Spreadsheet
    This reminds me of the way you tend to see DNA displayed. I wonder what sort of life form it really is?
    I was thinking the exact same thing.

    DNA electrophoresis

    [URL=http://img69.imageshack.us/i/20070706protgel2.jpg/][IMG]http://img69.imageshack.us/img69/6503/20070706protgel2.jpg[/IMG][/URL]



  • TRWTF is that someone is "working on a fix" for this problem. Haven't you heard of Team foundation server or any of the other millions of project management tools out there? BUY something for God's sake - don't spend $30,000 of a developer's time on this already stupid problem.



  • @jasmine2501 said:

    TRWTF is that someone is "working on a fix" for this problem. Haven't you heard of Team foundation server or any of the other millions of project management tools out there? BUY something for God's sake - don't spend $30,000 of a developer's time on this already stupid problem.
    We've looked at other software to do what we need, but our workflow is just too simple in some ways and too complex in others for most software to work well for us.

    We don't need something with a tons of features or lots of configuration.  The only reason things are as complex as they are now is because of the spreadsheets - take those away, and 90% of the complexity of the workflow goes away too.

    Fundamentally, though, it's only going to take me 2-3 weeks to build a simple system to replace all the various tracking spreadsheets.  I've already replaced all of our freelancer tracking spreadsheets with a simple system, and the improvements in productivity were huge.  Time spent on calling/tracking freelancers went from a few hours a day per employee per freelancer, down to less than 30 minutes.

    The users want something as simple as clicking a single button; that's something I can give them with a custom system that other software won't be able to.

    Coupled with major changes in our convoluted business processes, and we'd be in relatively good shape.



  • WHAT THE F*CK IS THAT AND WHY IS IT STILL ALIVE



  • @KrakenLover said:

    it's only going to take me 2-3 weeks to build a simple system to replace all the various tracking spreadsheets

    So, about $6500 worth of your time, just for the implementation phase? See where I'm going here? Even if you got a system with more than you needed, you would still save money on implementation. THEN, you could further save money on training because you will be using something WITH documentation already written, and probably tons of 'help' available on the web, forums, whatever (none of which your software will have). Then, you could save further money by firing some of the staff, who will inevitably hate the new system and refuse to learn it, and can then be dismissed for not knowing their jobs.



  • @jasmine2501 said:

    @KrakenLover said:
    it's only going to take me 2-3 weeks to build a simple system to replace all the various tracking spreadsheets

    So, about $6500 worth of your time, just for the implementation phase? See where I'm going here? Even if you got a system with more than you needed, you would still save money on implementation. THEN, you could further save money on training because you will be using something WITH documentation already written, and probably tons of 'help' available on the web, forums, whatever (none of which your software will have). Then, you could save further money by firing some of the staff, who will inevitably hate the new system and refuse to learn it, and can then be dismissed for not knowing their jobs.

    I'm with Jasmine. You gotta run the numbers.

    Look, everybody, everybody, thinks their business need is so unique that no existing product could possibly cover it. They're wrong. In fact, if anything I'd say the opposite: almost any product could be easily modified to meet your business need.



  • TRWTF is that ImageShack seems to have some sort of phobia against direct linking... they seem to have recently changed their "direct link" box to require signing in to an account to view, but they didn't actually get rid of the link in the text box, just made it read-only. A quick Inspect Element later and I was able to see the screenshot for it's full (glorious?) size.



  • @jasmine2501 said:

    @KrakenLover said:
    it's only going to take me 2-3 weeks to build a simple system to replace all the various tracking spreadsheets

    So, about $6500 worth of your time, just for the implementation phase? See where I'm going here? Even if you got a system with more than you needed, you would still save money on implementation. THEN, you could further save money on training because you will be using something WITH documentation already written, and probably tons of 'help' available on the web, forums, whatever (none of which your software will have). Then, you could save further money by firing some of the staff, who will inevitably hate the new system and refuse to learn it, and can then be dismissed for not knowing their jobs.

    First of all: I wish I made that much in a three week period.  I wish I made that much in a five week period.

    Secondly:  I have done enough research to know that an off the shelf piece of software is not going to do what we need out of the box, without a lot of customization, even if we change our business processes.  Which includes integrating with other systems (like our ERP software, legacy systems, and other custom systems).  I say that as both a developer, and as someone who actually uses our current system and understands the business requirements. And also as someone who has spent the past six months trying to figure out our legacy systems, having barely scratched the surface.

    In the time it would take me to evaluate all the software options available, hold the necessary meetings, build stakeholder buy-in, and talk to the vendor to figure out what system integration options are available, I could have designed, prototyped, and finalized a simple tracking software that does precisely what we need it to do and is easy to update and integrate with our existing and future systems - all at a lower cost to the company.

    That's not counting all the time I'd have to spend learning how to use the third-party system well enough to train the users, figuring out how to customize it, do user acceptance testing, deal with constant inquiries of "can we change this?" and my inevitable reply of "no, the software won't allow that" - or, god forbid, respond "you can use CrystalReports to do that".  And then there would be the endless cavalcade of highly paid consultants brought in to verify that everything will work adequately, and integrate properly, then the delays and contract extensions.

    Suffice to say, we've been down that road once or twice before.  It was costly, a huge time sink, and total overkill for our needs.

    Doing it myself saves so much time and money compared to the alternative.  If only because I can circumvent the bureaucracy that would be involved by using a third-party software.

    Adding yet another third-party black-box to our already convoluted mishmash of spreadsheets and systems, dealing with its idiosyncrasies, would just be one more nail in my coffin.

    @blakeyrat said:

    I'm with Jasmine. You gotta run the numbers.

    Look, everybody, everybody, thinks their business need is so unique that no existing product could possibly cover it. They're wrong. In fact, if anything I'd say the opposite: almost any product could be easily modified to meet your business need.

    Exactly.  Spot-on.

    Why even bother with getting a project management system?  MS Access is so flexible and versatile, I think we'll just use that instead.  Sure, it would take a few tweaks, and some spreadsheets here and there, maybe a little VB (I could probably get the end users to help with that, it's practically like writing in natural English), but eventually we'd get something that works great.  And to get the Mac users on it, we can set up a CITRIX server!  Man, that would be so awesome!

    *gunshot*



  • I used $75K/year and rounded the time up to one month, giving $6250 gross pay - there's also benefits they might be paying during that time, and developers always underestimate times, so the number is almost certainly higher than my estimate. And if you're not getting at least $75K to deal with this mess, you need to go somewhere else. You are clearly smart enough for it.


  • ♿ (Parody)

    @jasmine2501 said:

    I used $75K/year and rounded the time up to one month, giving $6250 gross pay - there's also benefits they might be paying during that time, and developers always underestimate times, so the number is almost certainly higher than my estimate.

    Also, depending on the nature of your job, the opportunity cost of not billing to a paying customer, where you can usually double to triple your salary once you figure in your benefits, G&A and profit. If your job is all in house stuff, then that's not so important.



  • @jasmine2501 said:

    I used $75K/year and rounded the time up to one month, giving $6250 gross pay - there's also benefits they might be paying during that time, and developers always underestimate times, so the number is almost certainly higher than my estimate. And if you're not getting at least $75K to deal with this mess, you need to go somewhere else. You are clearly smart enough for it.
    Well, I can't argue with your math.

    Right now I work two jobs just to get 73.5% of 75K.  One of which is at a public school (talk about wanting to crawl into a hole and die).

    This company has never really had a developer before.  And they still don't have an IT department.  Most of the IT stuff gets outsourced.  So, they have a tendency to undervalue IT people - they don't understand what it is we do, or why it's so hard to find someone who does it well. I'd like to think, though, that the hair I'm losing now is an investment in the company, and that I will get adequately rewarded eventually.  And, if not, I can move on.

    Right now, though, I'm willing to grin and bear it.  Make things better as best I can, and hopefully make a positive impact.  Despite all the complaining I do, and being underpaid, I actually really do like working here.

    And I get some measure of satisfaction from slaying krakens.  Feels really good to pack up some god awful spreadsheet or system, ship it off to some dark shelf at our back-up facility, and know that the replacement is going to make other people's lives a little less depressing.  If not my own.

    And, thanks.



  • @KrakenLover said:

    Which includes integrating with other systems (like our ERP software, legacy systems, and other custom systems).

    You just told us the whole company ran on Excel sheets.

    @KrakenLover said:

    In the time it would take me to evaluate all the software options available, hold the necessary meetings, build stakeholder buy-in, and talk to the vendor to figure out what system integration options are available, I could have designed, prototyped, and finalized a simple tracking software that does precisely what we need it to do and is easy to update and integrate with our existing and future systems - all at a lower cost to the company.

    Report back in 3 weeks, tell us how far you've gotten. Because I think you're either vastly, vastly overestimated your own ability, or vastly, vastly underestimated the complexity of the software required.

    To ask an obvious question, why do you think the time required to integrate with your ERP, legacy, and custom systems *changes* when you're developing the app, compared to using an off-the-shelf app? Why wouldn't that time be a constant?

    @KrakenLover said:

    Doing it myself saves so much time and money compared to the alternative.  If only because I can circumvent the bureaucracy that would be involved by using a third-party software.

    Why are "using third-party software" and "doing it myself" mutually-exclusive?

    @KrakenLover said:

    Adding yet another third-party black-box to our already convoluted mishmash of spreadsheets and systems, dealing with its idiosyncrasies, would just be one more nail in my coffin.

    Oh, so personal prejudice is the reasoning. Ok.

    @KrakenLover said:

    Exactly.  Spot-on.

    Why even bother with getting a project management system?  MS Access is so flexible and versatile, I think we'll just use that instead.

    You know what? If you had a copy of Access on every desktop, and a real MS SQL Server to provide the data store, yes, you could do this in MS Access. Yes, it is that flexible and versatile.

    @KrakenLover said:

    And to get the Mac users on it, we can set up a CITRIX server!  Man, that would be so awesome!

    How are the Mac users going to use the thing you're cooking up?



  • Well at least one of us got the joke :)



  • @blakeyrat said:

    Why even bother with getting a project management system?  MS Access is so flexible and versatile, I think we'll just use that instead.
    Why bother with Access? I'm sure you could do all this with an Excel spreadsheet...



  • @blakeyrat said:

    You just told us the whole company ran on Excel sheets.
    I don't recall ever typing that.  Unless I just forgot, and used an exaggeration some place.

    @blakeyrat said:

    Report back in 3 weeks, tell us how far you've gotten. Because I think you're either vastly, vastly overestimated your own ability, or vastly, vastly underestimated the complexity of the software required.

    If I have made a mistake and overestimated my abilities, or underestimated the task at hand, then that would hardly be unexpected.  I'm not saying that I am infallible, or that my plans and expectations are perfect.  I have failed plenty often enough to have learned that lesson.

    All I'm saying is that, at the end of the day, I trust my ability to listen to the users, understand their needs, and give them what they need (which isn't necessarily what they are asking for) more so than a vendor's ability to do that.

    @blakeyrat said:

    To ask an obvious question, why do you think the time required to integrate with your ERP, legacy, and custom systems *changes* when you're developing the app, compared to using an off-the-shelf app? Why wouldn't that time be a constant?
    Understanding the api of a system I designed gives me an advantage versus learning the api of two third-party systems and figuring out how to get them to talk to one another.

    Or, the alternative, doing it myself is cheaper than paying a vendor to integrate the systems. Set time = money, and there you go.

    Further - this project tracking system is only a small part of a much larger and more complex system I am also developing.  It's much easier for me to just plug in another module that I developed into the core system than it would be to try and integrate several different third-party software modules into it.

    Also, I don't trust the house-of-cards methodology to integrate disparate systems.  I want the final product to be simple, with a single interface, rather than a complex mess of systems developed by different companies using different technologies, strung together such that a single failure along the chain has negative repercussions elsewhere. Of course, that makes me distrustful of...pretty much every technology there is.

    And, consolidating a lot of functionality into a single system doesn't mitigate that kind of risk, it just sort of shifts it from external risk to internal risk.  But there's no such thing as an architecture with zero points of failure.

    @blakeyrat said:

    Why are "using third-party software" and "doing it myself" mutually-exclusive?
    I'm not quite sure what you mean by that.

    I will use some third-party libraries, and the system will have to integrate with other third-party systems.  So, it's not mutually exclusive in that sense.

    Or, if you are asking why an internally developed system can avoid the bureaucracy while a third-party system with similar functionality cannot, the answer is a combination of corporate politics and budget.

    @blakeyrat said:

    Oh, so personal prejudice is the reasoning. Ok.
    Yeah, I kind of have a thing against vendors charging an arm and a leg for lousy software that necessitates compromise after compromise, and doesn't really do what users want it to do.

    And I have a personal motivation in that I don't want to be on the receiving end of countless user complaints about the system and be left powerless to correct the issues.

    Granted, that then means that all user complaints I get will be pinned solely on me.

    So it goes.

    @blakeyrat said:

    How are the Mac users going to use the thing you're cooking up?
    It will be web-based.



  • @MiffTheFox said:

    TRWTF is that ImageShack seems to have some sort of phobia against direct linking... they seem to have recently changed their "direct link" box to require signing in to an account to view, but they didn't actually get rid of the link in the text box, just made it read-only. A quick Inspect Element later and I was able to see the screenshot for it's full (glorious?) size.

    Imageshack are quite the WTF. Every so often they do something to make viewing the "direct link" box more annoying, such as hiding it in a click-to-expand box, then revert it a bit later. At one point they had actually changed it to be a <script> tag that pulls some script from their server to display the image.


  • Considered Harmful

    @KrakenLover said:

    So it goes.


  • Discourse touched me in a no-no place

    @KrakenLover said:

    @blakeyrat said:

    To ask an obvious question, why do you think the time required to integrate with your ERP, legacy, and custom systems changes when you're developing the app, compared to using an off-the-shelf app? Why wouldn't that time be a constant?
    Understanding the api of a system I designed gives me an advantage versus learning the api of two third-party systems and figuring out how to get them to talk to one another.

    "Look now nice and square my newly designed wheel is; it won't roll down hills by itself like normal wheels..."

    Or, the alternative, doing it myself is cheaper than paying a vendor to integrate the systems. Set time = money, and there you go.
    Using 3rd party tools/apps != hiring 3rd party company employees to do the work.


  • @jasmine2501 said:

    I used $75K/year and rounded the time up to one month, giving $6250 gross pay - there's also benefits they might be paying during that time, and developers always underestimate times, so the number is almost certainly higher than my estimate. And if you're not getting at least $75K to deal with this mess, you need to go somewhere else. You are clearly smart enough for it.
     

    With such salary, i could buy a new house every 3 years, where do you live? :D



  • @tchize said:

    With such salary, i could buy a new house every 3 years, where do you live? :D
     

    The American balance of wage vs cost of living is slightly different than in most European countries, I believe.



  • @tchize said:

    With such salary, i could buy a new barraque à frites every 3 monthes, where do you live? :D
    FTFY



  • @KrakenLover said:

    it's only going to take me 2-3 weeks to build a simple system

     Or maybe 6 weeks?



  • @tchize said:

    @jasmine2501 said:

    I used $75K/year and rounded the time up to one month, giving $6250 gross pay - there's also benefits they might be paying during that time, and developers always underestimate times, so the number is almost certainly higher than my estimate. And if you're not getting at least $75K to deal with this mess, you need to go somewhere else. You are clearly smart enough for it.
     

    With such salary, i could buy a new house every 3 years, where do you live? :D

    Ah, if only that were true. You have to take into account that wherever that $75k is, the cost of living is equally as high. Plus, if you factor in the rule of thumb to not spend more than one week's pay on mortgage/rent, you are looking at a house that costs well over half a million dollars (not taking into account any sort of interest, taxes, insurance, HOA fees). This is not to mention things that go in the house, like furniture, blinds/window treatments, lawn care, etc and so on.

    OTOH, if you were to work remotely, that would be another story altogether. However, from my experience, I have yet to find a $75k/yr job that would allow anybody to work exclusively from home. But my area sucks for that sort of thing.



  • I see two front page histories,

    1. The Excel thingy as told by you
    2. The Replacement as told by your succesors

    Good luck



  • @PJH said:

    Or, the alternative, doing it myself is cheaper than paying a vendor to integrate the systems. Set time = money, and there you go.
    Using 3rd party tools/apps != hiring 3rd party company employees to do the work.

    That's the part I was trying to get across and obviously failed at after two tries. I understand Kraken doesn't like vendors, but what I don't get is why he thinks you need to hire vendors to use a 3rd party solution... even for commercial ones, the consulting is (almost always) optional.

    Look, take a pre-existing package that does 75% of what you need. There are tons available, and many are open source. (I wouldn't seriously recommend Access*, but you could look into Filemaker depending on your budget for the project. Filemaker's cross platform.)

    Of course, that doesn't change the fact that "doesn't like vendors" is a completely illogical way to go about producing software. You gotta run the numbers. There's no way I'd quote 3 weeks for an app like this. Especially if you're starting from scratch. It's going to take a week just to put all the data layers in-place, and that's being very very generous and assuming your existing third-party software isn't shit. Hell, it's going to take a solid week just to come up with the backup solution. 3 weeks is the "push a shoddy POS out the door and spend the next 2 months debugging it by the seat of my pants at 2:00 AM" quote.

    *) Note: Access is a fine product if used by people who know what the fuck they're doing. It's only shit when people use the built-in database as a multiuser database, something it is not and has never been. If you back Access with a real data store, it's a mostly WTF-free product that works fine. The reason I wouldn't recommend it here is your requirement to support Macs.



  • @blakeyrat said:

    3 weeks is the "push a shoddy POS out the door and spend the next 2 months debugging it by the seat of my pants at 2:00 AM" quote
    Isn't it the standard IT process ?



  • @PJH said:

    "Look now nice and square my newly designed wheel is; it won't roll down hills by itself like normal wheels..."
    Sometimes the wheel being square is part of the business requirements.  For better or worse.

    @blakeyrat said:

    3 weeks is the "push a shoddy POS out the door and spend the next 2 months debugging it by the seat of my pants at 2:00 AM" quote.
    More or less.

    This isn't the first spreadsheet "system" I've replaced in three weeks time or less at this company, with much fanfare from the users who reaped the benefits.  And it's far from the most complex. As of yet, the only bugs users have found are spelling errors.

    I'm not totally opposed to third-party software either, and I have used some to replace various other systems.  In my judgment, in this case, I don't think that it's appropriate.

    @serguey123 said:

    I see two front page histories,

    1. The Excel thingy as told by you
    2. The Replacement as told by your succesors
    That is a certainty.  I can hardly wait to see how I am characterized.

    @PJH said:

    Using 3rd party tools/apps != hiring 3rd party company employees to do the work.

    Typically, yes.  But me maintaining control over who does what work goes out the window when a third-party company is involved. I might end up with the task, or I might not.

    We are simultaneously migrating to a new ERP system.  Any purchase/acquisition of new systems that have to integrate with the ERP system will fall under the umbrella of that project, and our consultants/system-integrators will want a piece of the action (and will no doubt bring up their "solutions"); the implementation of the system I'm working on will be delayed and tied into the implementation of the ERP system.  Likewise, I would lose control of the scheduling system, as would the users - the users would then have little say over things while all the decisions are made by upper-management committees who are far removed from the daily realities of their employees.

    Having worked with this vendor before, and knowing what their idea of systems integration is (hint: "Sure, you can import anything with our super-flexible, state of the art, data import module!  Just put it into a spreadsheet!"), I am not willing to risk them screwing this up when all we need is something very simple.

    The department I'm in shields my projects from this vendor, as long as the scope is largely confined to my department and it doesn't involve the purchase of third-party systems.  The second I put in a purchase request for anything that's third-party, our IT services vendor gets involved, cross-departmental meetings are held, and someone will involve the above mentioned contracted systems vendor.

    Like I said before, there are corporate politics involved, and a lot of people have a lot of interest in either grabbing political capital or money, as the case may be. And in that prestige/cash-grab, the needs of the users will be forgotten.

    @blakeyrat said:

    Of course, that doesn't change the fact that "doesn't like vendors" is a completely illogical way to go about producing software.
    In this case, it's "doesn't like a specific vendor with whom we have unfortunately signed a contract". If my superiors, and co-workers, didn't feel the same way as I do, then the vendor would have been given this project, and not me.

    And, originally, they were given this project.  They drew it our over the course of a year, and screwed it up so badly that we broke the contract we had for that project.

    No one, most especially the users who suffered through a year of training on the vendor's convoluted rube-goldberg-like "system" before I got here, wants to see them wedge their way into the projects I'm working on.



  • @KrakenLover said:

    We are simultaneously migrating to a new ERP system.  Any purchase/acquisition of new systems that have to integrate with the ERP system will fall under the umbrella of that project, and our consultants/system-integrators will want a piece of the action (and will no doubt bring up their "solutions"); the implementation of the system I'm working on will be delayed and tied into the implementation of the ERP system.  Likewise, I would lose control of the scheduling system, as would the users - the users would then have little say over things while all the decisions are made by upper-management committees who are far removed from the daily realities of their employees.

    Having worked with this vendor before, and knowing what their idea of systems integration is (hint: "Sure, you can import anything with our super-flexible, state of the art, data import module!  Just put it into a spreadsheet!"), I am not willing to risk them screwing this up when all we need is something very simple.

    The department I'm in shields my projects from this vendor, as long as the scope is largely confined to my department and it doesn't involve the purchase of third-party systems.  The second I put in a purchase request for anything that's third-party, our IT services vendor gets involved, cross-departmental meetings are held, and someone will involve the above mentioned contracted systems vendor.

    Like I said before, there are corporate politics involved, and a lot of people have a lot of interest in either grabbing political capital or money, as the case may be. And in that prestige/cash-grab, the needs of the users will be forgotten.

    @blakeyrat said:

    Of course, that doesn't change the fact that
    "doesn't like vendors" is a completely illogical way to go about
    producing software.
    In this case, it's "doesn't like a specific vendor with whom we have unfortunately signed a contract". If my superiors, and co-workers, didn't feel the same way as I do, then the vendor would have been given this project, and not me.

    And, originally, they were given this project.  They drew it our over the course of a year, and screwed it up so badly that we broke the contract we had for that project.

    No one, most especially the users who suffered through a year of training on the vendor's convoluted rube-goldberg-like "system" before I got here, wants to see them wedge their way into the projects I'm working on.

    ... why do you work for this company again?



  • My employer used to have a labor management system (to track what happens when for scheduling purposes) created by the national office of the coop we're part of. The pair who created it were from a different king of retail company, where there was a better understanding of the value of IT, and they had little support in creating Version One. Which is to say, their budget for developers was $0.00, and they ended up with the person from the secretarial pool who claimed to know how to do marcros in Excel.

    What they ended up with was an Excel spreadsheet that was too big to email (and we run our own mail server), and literally had minimum hardware requirements (2 GB of RAM, IIRC, and this was long enough ago that most of our beefier office machines were running 512 MB).



  • @blakeyrat said:

    ... why do you work for this company again?
    Probably because I'm a little naive, and think I can help make things better.

    Also - my other job is working at a public high school.  Comapred to that, I'm willing to put up with just about anything.

    @taustin said:

    My employer used to have a labor management system (to track what happens when for scheduling purposes) created by the national office of the coop we're part of. The pair who created it were from a different king of retail company, where there was a better understanding of the value of IT, and they had little support in creating Version One. Which is to say, their budget for developers was $0.00, and they ended up with the person from the secretarial pool who claimed to know how to do marcros in Excel.

    What they ended up with was an Excel spreadsheet that was too big to email (and we run our own mail server), and literally had minimum hardware requirements (2 GB of RAM, IIRC, and this was long enough ago that most of our beefier office machines were running 512 MB).

    I feel your pain.

    We have lots of "networked" spreadsheets - sent either via email or sneaker-net. It's really scary when you have one that won't even fit on one of the official company flash-drives (which, btw, are 1 gig).



  • @KrakenLover said:

    The solution, of course, is not to rethink the whole tracking methodology, but to add another layer of tracking spreadsheets that record more detailed tracking data.

     

     

    Wonderful thing about excel, as opposed to a (presumably more robust) actual project tracking system:  Updates flow through an excel spreadsheet system from one form to another without logging/history/undo   Your project is behind schedule?  Too many hours worked?  Too many bugs?  Just change a few numbers, wait for the new printouts, and -hey- you're back on track.

    Seriously, if the company is making budgeting decisions based on project progress, I'd think they'd want to ensure that the numbers they're looking at are correct, not the result of some manager somewhere fat-fingering and changing last month's values by mistake.



  • I've been somewhere that was pretty much the same.

    "We need to keep track of projects" -> Spreadsheet.
    "We need to keep track of the unique ids we give projects" -> Another spreadsheet
    "We need to keep track of the latest version of software for each project" -> Another spreadsheet (one project per row)
    ...
    "We need to access this data company wide" -> Shared spreadsheet on global share (anyone could access)
    ...
    years pass and the spreadsheet becomes corrupted and fixed (retyped from printed copies) several times
    ...
    "Must be more reliable and we need to store much more data" -> Instructions added to first page requesting "When you have finished modifying the document, increment version number and save under a different name"
    ...
    eventually when they wanted much more data and were struggling to parse it -> Shared access database

    I think it's still an MDB file now, the excuse being they have "no time" to implement a proper system. 

     

     dogbrags :Spreadsheets such as Excel/Open Office DO have logging capabilities if they are a shared document. They'll even highlight what has been changed. So if someone edited the wrong project, you can discover it and revert the change. You even know the user that made the change.

     

     



  • @dogbrags said:

    Seriously, if the company is making budgeting decisions based on project progress, I'd think they'd want to ensure that the numbers they're looking at are correct, not the result of some manager somewhere fat-fingering and changing last month's values by mistake.
    Yeah; they are doing budgeting, and prioritization based on this.  And it is far from correct.  Right now, the Access DB reports are transposing columns.  Among the affected fields is the "On hold date" column getting transposed with the "initial phase complete" date.

    Debugging several thousand lines of 20 year old MS Access SQL written by people with little or no technical experience?  Fun times.

     @Mole:   Sounds like you've worked here before. That describes things to a tee.

    @Mole said:

    dogbrags :Spreadsheets such as Excel/Open Office DO have logging capabilities if they are a shared document. They'll even highlight what has been changed. So if someone edited the wrong project, you can discover it and revert the change. You even know the user that made the change.
    Yes - assuming that the people making the changes are changing their own projects, and there are only a handful of spreadsheets accessible to everyone.

    The way we do things, project managers keep their own spreadsheets in each project directory.  Then, periodically, they type out an email and send it to their team leader.  The team leader then manually types the info from the email into their own tracking sheets.  Then the team leaders type out an email and send it to the office administrator.  The office administrator then manually updates the master tracking sheet and Access database.  The office administrator then runs an Access report and saves it as a PDF which then gets printed out and emailed around the company.

    Numbers can be changed or fudged at any point along those lines, with no track record of who did what.  The same goes for mistakes - there's a lot of manual retyping going on, and format changes. Really, I think mistakes are more common than intentional deception here.



  • @KrakenLover said:

    The way we do things, project managers keep their own spreadsheets in each project directory.  Then, periodically, they type out an email and send it to their team leader.  The team leader then manually types the info from the email into their own tracking sheets.  Then the team leaders type out an email and send it to the office administrator.  The office administrator then manually updates the master tracking sheet and Access database.  The office administrator then runs an Access report and saves it as a PDF which then gets printed out and emailed around the company.
    Obviously without knowing the details, I can't say much about it, but it sounds to me like your problem is more that you have a bad Excel workflow than using Excel itself. Well, OK, Excel isn't the best choice, but the problems sound like they're more to do with really messed-up implementation than anything fundamental. Wouldn't it not be easier just to link various worksheets together, and let them update each other automatically?



  • @intertravel said:

    Obviously without knowing the details, I can't say much about it, but it sounds to me like your problem is more that you have a bad Excel workflow than using Excel itself. Well, OK, Excel isn't the best choice, but the problems sound like they're more to do with really messed-up implementation than anything fundamental. Wouldn't it not be easier just to link various worksheets together, and let them update each other automatically?
    Yes, the implementation is very poor to. Believe me, there's plenty of WTF sprinkled on our business processes.

    If there were another software or tool as ubiquitous as Excel, then the company would have abused that instead.  If half the company didn't use Macs, then I don't doubt that I'd be dealing with MS Access way more than I am now.

    One of the things I'm working on is changing the way we do things in general.  Not just replacing inadequate technology, but throwing out old processes that just don't make sense anymore, and developing new ones that are efficient and add value rather than being an obstacle to getting work done quickly & well.

    That said, there are alternatives to developing a custom system. But I'm loath to encourage the use of Excel in this manner because it's just going to send the wrong message; and won't help my co-workers to think outside the box and realize that there are better solutions out there than networked spreadsheets.



  • @KrakenLover said:

    there are better solutions out there than networked spreadsheets

    You... you take that back!



  • @blakeyrat said:

    I'm with Jasmine. You gotta run the numbers.

    Look, everybody, everybody, thinks their business need is so unique that no existing product could possibly cover it. They're wrong. In fact, if anything I'd say the opposite: almost any product could be easily modified to meet your business need.

    Like Excel? :D


  • @Master Chief said:

    @blakeyrat said:
    I'm with Jasmine. You gotta run the numbers.

    Look, everybody, everybody, thinks their business need is so unique that no existing product could possibly cover it. They're wrong. In fact, if anything I'd say the opposite: almost any product could be easily modified to meet your business need.

    Like Excel? :D

    There's not a lot you can't do with Excel, given sufficient incentive. Almost all of it would be better done another way, but Excel can do it. It's quite disturbing that quite a few of the business apps I've run across are actually even worse than hacking something up in Excel would be, particularly where the business is only using a small subset of the application's features.

    Whilst complaining about/advising upgrade of the ineffable shitness of the case management software one place I worked, I said something like 'I could do better in Excel'. Obviously, they laughed. Seemed an interesting thing to try and do, though, so in my break that day I spent half an hour roughing out an Excel system. Boss had a look and said 'oh, you're right, it is better in Excel - how long would it take you to finish this?'

    After beating my head on the desk for a minute or two, I persuaded him that buying a decent case management system was a better idea.

    @KrakenLover said:

    One of the things I'm working on is changing the way we do things in general.  Not just replacing inadequate technology, but throwing out old processes that just don't make sense anymore, and developing new ones that are efficient and add value rather than being an obstacle to getting work done quickly & well.

    That said, there are alternatives to developing a custom system. But I'm loath to encourage the use of Excel in this manner because it's just going to send the wrong message; and won't help my co-workers to think outside the box and realize that there are better solutions out there than networked spreadsheets.

    That makes sense to me. (And can I just say that unlike some others here, I can understand why you'd want to work there - sounds interesting.) I find the idea quite amusing, though, because there have been a few times I've done the exact opposite - get rid of an overly heavyweight app, server to run it, maybe a Citrix (spit) server to go with it, and so-on, in favour of a simple set of linked spreadsheets totalling a few megs.

    (Not that I'm saying you app is overly heavyweight, or anything like that. But the ones I changed were.)



  • @intertravel said:

    There's not a lot you can't do with Excel, given sufficient incentive. Almost all of it would be better done another way, but Excel can do it. It's quite disturbing that quite a few of the business apps I've run across are actually even worse than hacking something up in Excel would be, particularly where the business is only using a small subset of the application's features.

    Excel is like a Swiss army knife.   It has enough features and attachments to do just about anything.  That said, you wouldn't use the little saw to cut down a tree.  At least a person with common sense wouldn't.  But, when that Swiss army knife is the only tool you know how to use, and you aren't even aware that there exists things like axes and chainsaws, then you can think up some pretty creative ways to use what you have to get done what you need to do.

    So, while I might prefer a single chainsaw for cutting down a given tree, people who only have that Swiss army knife at their disposal, or have never even heard of a chainsaw, will probably think up a way to use twenty or thirty Swiss army knives, along with some duct tape, to cut down the same tree.

    Of course, a chainsaw is a chainsaw.  You wouldn't use it to scale a small fish, or unscrew something (though, some people do certainly try).  Sometimes that Swiss army knife is just perfect for the task at hand, even if you're only using two or three of its little gizmos.

    The trick is knowing which tool is most appropriate for which circumstance.

    People who have never solved a problem using anything but Excel don't even realize that there are other options out there.  Often I don't just think it's just a matter of "if you have a hammer, everything looks like a nail", but "if you have a hammer, everything looks like a nail, and you are completely blind to all other tools in existence - even when taking a full tour of Home Depot".

    @intertravel said:

    That makes sense to me. (And can I just say that unlike some others here, I can understand why you'd want to work there - sounds interesting.) I find the idea quite amusing, though, because there have been a few times I've done the exact opposite - get rid of an overly heavyweight app, server to run it, maybe a Citrix (spit) server to go with it, and so-on, in favour of a simple set of linked spreadsheets totalling a few megs.

    (Not that I'm saying you app is overly heavyweight, or anything like that. But the ones I changed were.)

    It all comes down to using the right tool for the job.



  • @KrakenLover said:

    Often I don't just think it's just a matter of "if you have a hammer, everything looks like a nail", but "if you have a hammer, everything looks like a nail, and you are completely blind to all other tools in existence - even when taking a full tour of Home Depot".

    I think often it's simpler than that: 'you have a hammer'.

    @KrakenLover said:

    It all comes down to using the right tool for the job.

    Absolutely. Or a hammer, because hammering is fun :)


  • Discourse touched me in a no-no place

    @KrakenLover said:

    The trick is knowing which tool is most appropriate for which
    circumstance.

    No



    It's not.



    We knew /that/ 5 iterations ago. The trick is explaining why, to others, their pet software is wrong for the circumstance. And geting that explanation to hold through the next 5 iterations.



  • @intertravel said:

    @KrakenLover said:
    Often I don't just think it's just a matter of "if you have a hammer, everything looks like a nail", but "if you have a hammer, everything looks like a nail, and you are completely blind to all other tools in existence - even when taking a full tour of Home Depot".
    I think often it's simpler than that: 'you have a hammer'.
    That's a lot of it, of course. Most office workers have a hammer given to them when they start work; if they want any other tool, they have to be able to justify the purchase and wait for it to go through channels. In the meantime, there's this hammer just sitting there...
    @intertravel said:
    @KrakenLover said:
    It all comes down to using the right tool for the job.
    Absolutely. Or a hammer, because hammering is fun :)
    QFT.



  • @Scarlet Manuka said:

    Filed under: Hammer time

    Stop!

     



  • @blakeyrat said:

    @KrakenLover said:
    Which includes integrating with other systems (like our ERP software, legacy systems, and other custom systems).

    You just told us the whole company ran on Excel sheets.

    @KrakenLover said:

    In the time it would take me to evaluate all the software options available, hold the necessary meetings, build stakeholder buy-in, and talk to the vendor to figure out what system integration options are available, I could have designed, prototyped, and finalized a simple tracking software that does precisely what we need it to do and is easy to update and integrate with our existing and future systems - all at a lower cost to the company.

    Report back in 3 weeks, tell us how far you've gotten. Because I think you're either vastly, vastly overestimated your own ability, or vastly, vastly underestimated the complexity of the software required.

    To ask an obvious question, why do you think the time required to integrate with your ERP, legacy, and custom systems *changes* when you're developing the app, compared to using an off-the-shelf app? Why wouldn't that time be a constant?

    It's been 3 weeks!! Time to check back with KrakenLover and see how far he's gotten!


Log in to reply