State of the profession

Dick M asks if we (Office developers) are a dying breed. At least I think he’s asking - he could be telling us our profession is wilting away. Whatever, I agree. Completely.

Office developer, Excel developer, business developer, these mean pretty much the same thing to me, its a person nearer the business than a traditional IS department dev. They probably use a range of tools/platforms, they mainly target desktop rather than server, Excel usually features as at least part of the interface (or calc engine), source data may be from a server database, but working data is likely to be in jet. Glue code may be VB6 or .net but most likely VBA. The best differentiators (from mainstream devs) are around business acumen rather than technical tool choice.

Anyway Dick mentions a few items, here are my thoughts:

Value for work - Bad and getting worse, its getting so only City institutions understand the value of Excel, which is bizarre considering their normal sheep like tendencies. ( I suppose they are flocking round each other).

Respect - Bad and getting worse, for years I was embarrassed to admit I did a lot of work in Excel, as ‘real devs don’t touch Excel - hobbyists only’. Then as Eusprig gathered momentum I got a bit more confident, but I think that quality/error movement has stalled a little now. So back to claiming to be a mainstream dev for me I reckon.

Continuity - Bad and getting worse, I used to coach client staff in what I was doing in Excel/Access/VBA etc. These days they turn their noses up, happy to pick up some SQL or C# but no Excel/VBA thank you. I just don’t see many people rushing into our technologies.

Conclusion - yeah we’re proper fooked. Unless MS are going to wake up to their most valuable asset - MS Office on an MS Windows desktop. By wake up I mean the hand in pocket wake up, not the cheap soundbite one thats been (pointlessly) running (well walking) for as long as I can remember.

I don’t think us devs will be the losers we’ll just move to techs where the opportunities are better. The real losers will be MS as customers move to other platforms that have an apparently better cost/benefit story, but only because they are unaware of much of the benefit of the MS Office platform. [Ignoring for the time being the cost/no benefit fubar known as the 2007 UI shuffle]

The other losers of course will be businesses far and wide that have to wait for their IS departments to implement their grossly over engineered, over priced, over due, big iron monsters, just in time to realise the business requirement has moved on.

The winners are of course the IS departments who gain control, power and loads of low pressure work. If you know you will release straight into retirement (rather than production) there are lots of unpleasant things you can forgo, like testing, and documentation for example.

Is this how you see things?

or are you seeing different trends?

How do you think things will pan out in 3-5 years?

Do you have an escape plan?

Please comment here or on Dicks post

cheers

Simon

Excel consulting firms

Where are they?
Where are the multi person Excel, or other spreadsheet, consulting/development firms?
If this technology is as important and useful as we think it is, and its hardly new (mature I think is the correct terminology (one step before legacy)), then why aren’t there more/any 10/20/100/200 person specialist firms?
Sharepoint has them, why not us?
(Is it because that is a ‘professional’ tech and Excel isn’t?;-))
Its like Excel dev is locked in the artisan phase and never going to move towards a more ‘engineered’ approach.
I’m not overly worried from a quality POV as I think if you get the right artisan you will get quality far beyond some tick list based ‘way’.
Its credibility that worries me, I think a few formalised big consulting firm standards would give the whole market a bit more clout. Even if they did conflict massively.
So where are they, and why aren’t there more?
cheers
Simon

First thing

Whats the very first thing you look at/for when you open a workbook from someone else? (that you trust not to be malicious)

And whats the first thing you look at/for when looking at someone elses code?

For spreadsheets its neatness/layout of whatever opens
too neat = gold plating/time wasting
too scruffy = slack/ little care

(Of course the ‘correct’ level of neatness is the sort of stuff I produce ;-)). I’m thinking use of formatting, use of number formatting, use of space etc.

I’d like to pretend thats its some kind of rational measure (file properties maybe?), but really its just a judgement on the initial visual appeal.

For (VB/VBA) code it is Option Explicit - missing = cowboy everytime. If its a snippet/function then I look to see if parameters variables and return are typed (as in Dim … as …., rather than dim x, y, z)

What about you?
one only for each category.
And is it relative to your own work? or do you think its an absolute measure?
cheers
Simon

Software developers are patronising

I just switched on my pc and went and brewed up. It turned itself off - thanks.

I read some software developer guidelines years ago that the default action of any dialog should be to leave the computer as is. Great! that saves idiots from accidentally destroying their computer. But it also means the default action is to ignore the users request. That seems pretty dumb/arrogant to me.

The Apple approach (as I understand it) is to let the user do whatever they want, but make it easy to undo. That seems to make more sense than to make it hard for them ‘for their own good’.

I did try turning off the ‘are you sure you want to delete that file’ warning, as I know I can get it back from the re-cycle bin, but I like the comfort of clicking Yes (without thinking) to the warning so I put it back.

I find that I turn off many of the Excel warnings, as they are covered by undo anyway. Things like ‘are you sure you want to overwrite those cells’ when pasting is turned off (along with all similar warnings) as I know I can either undo or revert to a saved version if things go badly wrong.

Do you agree that some of this stuff is patronising?

Do you turn off the warnings (Excel and elsewhere)?

What do you do with the systems you build for others?

cheers

Simon

Eusprig topic suggestions

I wasn’t planning on submitting a paper for consideration at this years Eusprig. But now I’m thinking I should. The deadline is looming however and I havent really got an obvious topic, thats where you come in…

The conf title is ‘in pursuit of spreadsheet excellence’ and papers need to relate to that somehow.

Last year we did spreadsheet hell, covering some of the cultural issues around corporate spreadsheet use.

The year before I did the process I use for reviewing client spreadsheets

The year before that I did some of the strengths and weakness of spreadsheets compared to other technologies.

[one possible one is User defined functions in XLM or something similar around function definitions]

So my question is, is there a topic you think I should cover, or that we should cover as a group?

Many of the Eusprig prior papers have been archived on-line here.

Please leave any suggested topic areas as comments, thx

(anything that incorporates my new logo suggestion is especially welcome!)

cheers

Simon

Why VBA is so rubbish

The last VBA post may have given the impression I’m some kind of VBA fanboi. I think it has its place, but it’s far from perfect. I put my definition in that post so I wont repeat it here.

I think its important to keep in context of what is out there live in the real world now, so I’m not going to suggest another language is better. In pure terms python (or your personal favourite language) maybe a better language for users to do basic automation etc. The reality is there is a lot of VBA code around, and more gets written everyday. I think the language would benefit from a few modern features, perhaps to support OO more fully, but not at the cost of backwards compatibility.

I’m also going to skirt around the issue of user competence. You get crappy code in every language, some languages may encourage it more, or may be more attractive, or more accessible, to developers of limited skill and experience.

It has been suggested that giving something as powerful as Excel/VBA to an unskilled user is the equivalent of giving a loaded rifle to a child. The reality, I would suggest, is a little less dramatic, as far as I know, no one has ever been killed by a poor quality spreadsheet, and very few companies have gone bust because of one. Yes SocGen just lost a stack of cash, but a person did that by trading badly, not a spreadsheet.

VBAs biggest fault?

I’d say its too easy to go beyond its design envelope.

For simple automating Excel and standalone complex worksheet functions (eg: the stuff Mike talks about) I think it is great. Its easy to slip from there to 10/12/14 KLOC systems, at which point I think VBA can work against you.

The other side to that argument is perhaps there is no realistic migration path, or no realistic alternative. That may or may not be true, but either way is not a fault of VBA.

Second - The editor. Frankly its so arcane I think Integrated Development Environment is a bit rich. If this just got updated to the VB6 one (which has more visible object model) add-in devs would fix everything else. Ideally though I’d like a VSTA style editor that writes proper VBA, or something completely compatible, in Excel 14.

That would also fix my third biggest fault - Microsofts lack of public love for VBA. Many devs are worried about its future, and hollow verbal assurances don’t cut it. A significant investment in upgrading the IDE, or replacing it with the VSTA one would settle the uncertainties for many.

Speeding up the call interface would be excellent too. As would allowing VBA UDFs to participate in multi-threaded calculation.

VBA’s reputation as a meddlers monster creation system (/toy language) hardly makes it the sort of thing you want to own up to at a party. The political correctness bureau solution to that is the change the name (eg ‘disabled’ went to ‘people with disabilities’ now moving to ‘differently abled people’) in general the ‘name’ gets longer as they use more circuitous language. So perhaps renaming VBA to something like AAS (Application Automation System) would fix any stigma? (of course it would have to be MAAS or MOCAAS (Microsoft Office Client AAS)). You get the idea.

Any other suggestions? the ruder the better.

What would be your top 2 or 3 weaknesses of VBA?

cheers

Simon

In Pursuit of Spreadsheet Excellence

The 2008 Eusprig conference is coming up in July. There is a current call for papers, the deadline for which is the end of March. Full info here.

If you are wondering about submitting a paper, but aren’t sure, feel free to email me for an informal chat. I’m nothing to do with paper selection, but I’ll happily discuss my experiences as a speaker.

Personally I think a few more papers from people who earn their living working with spreadsheets, talking about real world quality would be very welcome.

wtf is ‘Spreadsheet excellence’ anyway?

I reckon its the skilled application of suitable quality spreadsheet based components to appropriate issues, and the effective management of their lifecycle.

What does it mean to you?

Vista SP1

Vista SP1 seems like a bit of a difficult birth, but if you have it and your Office apps start crashing try disabling all the add-ins. This is Word 2007 related, has anyone had Excel problems?

cheers

Simon

Whats so good about VBA?

A (very valid) question from Johan.

Here is why I think VBA is so great, but first lets try and define VBA.

A tightly integrated application automation system that allows professionals from non software development disciplines to enhance their organisations use of the host application. With justifiable levels of training costs and tooling costs.

And secondly, lets add some context - we don’t develop in a vacuum so we should not review our tools in one.

Its 2008, VBA was integrated into Excel in v5 in 1995, last notable update was 97 really when we got the current IDE. Prior to VBA many orgs has dipped a toe in the macro waters with XLM and Lotus macros. Many organisations have over 10 years worth of live VBA code. And coders with 10 or more years experience solving that orgs problems with that toolset.

Software vendors desperate to sell us their latest version have almost convinced us that ‘legacy’ means ‘old useless shite that should be thrown out’. But legacy also means ’something of value handed down from ancestors’. What better legacy than tested, proven, working code, no matter how out of fashion the language might be? or how ‘not to your style’ the code might be?

I’ll be frank, I don’t much care for BASIC derived languages, I prefer C based ones. VBA has a few language features missing, and a few backwards compatibility ones I’d happily see removed. It also has the odd inconsistency, and a few awkward twists (If not (x is nothing) then... springs to mind). VBA is more, much more than a language though, its a complete system in partnership with the host app, with history.

‘Do you like the whole VBA story or not?’ boils down to do you savour proven, working code? or do you like new stuff? Do you like safe and certain? or do you like unnecessary risk and waste?

VBA has a major entry barrier for someone from a user background (I mean business here, but also science, engineering etc - anything except sw dev I guess). But once over that barrier is a gentle learning curve through to extremely high levels of productivity. Object Oriented approaches represent an even bigger barrier to that target group in my opinion, with limited payback in that problem domain.

Excel is a functional tool, its users understand functions - they write them in almost every cell. Very few use classes, objects make no sense to Excel users. Classes are considered an advanced technique in Excel VBA, not because they are difficult, because they are an unusual way for a spreadsheeter to go about solving the sorts of problems they solve with spreadsheets.

Spreadsheeters email their spreadsheets around, this approach is absolutely proven to work in (and out of) organisations of all sizes all over the world in all industries. VBA supports this zero deployment in almost all cases. Most spreadsheet users have never heard the expression ‘deployment’ that is how ‘zero touch’ VBA is. (many have never heard of testing either but that is a separate discussion)

The Excel COM object model is an ideal target for VBA, optional parameters, parameterised properties etc are all fine. Unlike C# for example, which is from a whole other era, designed from the ground up to solve a whole other type of problem. Should Excel get a C#/manged code friendly object model then that would even things up a little. But by then there might be 15 or 20 years (or more??) of VBA legacy code to compete with.

Whats so good about VBA?, is like saying whats so good about driving on the left? honestly? it would be better for a whole host of reasons if we swapped and got in line with all the new world countries. But who, when, how and where could this be done? and who will pay, and when would they get payback (hint: never ever ever).

Whats so good about VBA? its here, now, in widespread use. Perhaps the correct question is what should an organisation that is already heavily invested in VBA (most are) do now in 2008?

Of course theres lots to not like about it too, but thats a topic for another post. The point of this post is that I don’t see anything currently better than VBA for most of the stuff that VBA is used for, and for the folks that use it.

Maybe the VBA argument comes down to: SW devs will always hate it because its an impure bastardisation of a beginners language. Business users will prefer it to the other available options because they can leverage existing investments, so it represents the best value for money.

Try this test: ribbon v commandbars? .net v VBA? either you like new stuff or you like existing proven stuff.

What have I missed? anything to add? anything you disagree with?

cheers

Simon

Ribbon FAIL

Sorry I couldnt resist.

I couldn’t think how to show the biggest fail - shuffling the commands to alienate experienced users. I can’t comment on the apparent programmability fail. the picture shows a mystery fail and 2 ridiculous size fails.

I was going to change it back to the more familiar blue scheme, but I couldn’t find the command. Don’t bother telling me, I don’t need to know.

;-)

simon