An update on progress

[Shian]

Yes, sure...

Tell me please. You as a proffessional developer, if you start on a project and you've worked on it for a few days? weeks? are you able to say something / anything about when you will be finished?

Like it will take me at least another 2, 3, 4 weeks or months or do you know afer some time it will take more time, or maybe it is totally impossible?

Or do you say to your costumer, hey, it will be done when I'm done and that's all you will hear about it from me?

Programming is picky and alot like building with legos.

If you put one piece on, another falls off. So, if you're putting something together, there's not telling how many other pieces will fall off, what other bits of code it will affect, and the amount of time it will take to put them back together. That is why timeframes are usually not able to be given until the project is nearly done.

[Velea]

Yes, sure...

Tell me please. You as a proffessional developer, if you start on a project and you've worked on it for a few days? weeks? are you able to say something / anything about when you will be finished?

Like it will take me at least another 2, 3, 4 weeks or months or do you know afer some time it will take more time, or maybe it is totally impossible?

Or do you say to your costumer, hey, it will be done when I'm done and that's all you will hear about it from me?

I'm not a software developer, but I can answer that... Anyone who's ever done anything with a computer can answer that....

No. You can't say when something will be finished when it comes to a computer. You can't even say when it will be ready when it comes to a computer. You can't even say when you hope it will be ready, in fact. Because, you see, when Murphy first started making and programming computers, he make sure that every "Law" he ever stated was programmed so deeply into the foundation of a computer's logic that the one reliable thing you can say about a computer is that if something can go wrong with it, it will, and at the worse possible moment.

Rule number one of any computer class anywhere, from kindergarten right up to doctoral class, should be "Never ever give a date for delivery."

[tramsan]

Tell me please. You as a proffessional developer, if you start on a project and you've worked on it for a few days? weeks? are you able to say something / anything about when you will be finished?

Yes, but they invariably turn out wrong, and the same goes for any other developer I've worked with. Software development is a horrible business. Everyone seems to think it's some factory fabrication method, where you just put all the pieces together, slap a sticker on it and then ship it - it couldn't be any further from the truth. It's part engineering, part creative art and part just slugging it out with the system until it works.

The closest I've ever been to meeting a big deadline was 5 hours into the day I was supposed to be done - and by then I hadn't gone home since getting to work the day before. The system still got rejected by the client, because "it was not what we expected" (Fine, then why the hell did you give me a *gif* of the layout that your "web designer" had created, that you demanded be followed to the pixel?)

I'm as impatient as any of you to get my old characters back - I have about three years of investment gathering dust in the unity database, too - but on this particular issue I'm willing to give VI all the leeway they need, because I believe that they can do it. (As opposed to a certain former "owner")

[Danna Firesoul]

Yes, sure...

Tell me please. You as a proffessional developer, if you start on a project and you've worked on it for a few days? weeks? are you able to say something / anything about when you will be finished?

Like it will take me at least another 2, 3, 4 weeks or months or do you know afer some time it will take more time, or maybe it is totally impossible?

Or do you say to your costumer, hey, it will be done when I'm done and that's all you will hear about it from me?

Well, the other posters have handled this pretty thoroughly, but i figured I'd just confirm it. There's a trick to giving project estimates when you're a programmer. 1) Estimate how much work you think the project is. 2) Now estimate how much work it would be if everything you can think of goes wrong. 3) Divide that into 8 hours days. 4) Double that, because you're lucky if you get 4 heads down hours of work done on a project a day. 5) Lie to your manager, because you know something unexpected will happen. 6) Pray that the things you didn't think of that go wrong balance out the things you did think of that didn't.

It works about half the time. If you have a good manager, he understands that an estimate is exactly that, an estimate. If you have a bad manager, you get called out on the carpet for going over projection.

And, of course, there are the projects where it takes 2-3 days to program, and 2-3 months to figure out how to do it. And you have no idea how long any of it is going to take until you're almost done. There are more of those than you'd think.

Development is something like writing ("So, Mr. Dickens, when can we expect your next literary masterpiece?"), and something like invention, ("So Mr. Tesla, when can we expect your next revolutionary invention?"). But, like artisans with patrons back in the middle ages, we must somehow contrive to create on demand. But when we're honest, we have to admit that timetables and development are only casual acquaintances at best.

[Goriax]

1) Estimate how much work you think the project is. 2) Now estimate how much work it would be if everything you can think of goes wrong. 3) Divide that into 8 hours days. 4) Double that, because you're lucky if you get 4 heads down hours of work done on a project a day. 5) Lie to your manager, because you know something unexpected will happen. 6) Pray that the things you didn't think of that go wrong balance out the things you did think of that didn't.

It works about half the time. If you have a good manager, he understands that an estimate is exactly that, an estimate. If you have a bad manager, you get called out on the carpet for going over projection.

You only double the time? You like to live dangerously. I triple it. For me it isn't so much only being able to get 4 hours work per day due to other crap (4 hours of crap, then 8 hours of work); it's more a case that software development is ruled by Murphy's Law like nothing else.

If you've got one of those managers who don't understand these things you can take some small comfort in the fact that he's eventually going to go down. Hard. Because he's managing software development while knowing nothing about it.

Yea, people who don't do software development for a living seem to think that writing software is like building houses; where every program is very much like the one before and any halfway competant developer should be able to accurately predict how much time it'll take. It isn't. Not even close.

If it were me I wouldn't say what was coming until it was done, tested, and ready to go. Even then I'd be nervous.

If the player base were forgiving about features which don't make it to release then one could be more open. Developers have tried that. What seems to happen when they try to be open and keep players informed is that sooner or later something doesn't work out (pretty much guaranteed). Then some players start ranting about broken promises. Apparently the mere mention that you are working on something is construed as a solemn promise. In that kind of atmosphere it's best to keep closed mouthed and say as little as possible.

[Helious]

Oh heck can't resist with an old programming story myself

20 years ago .... yes there were computers back then .. i used to be a programmer (i gave it up because i wanted a life (i forget what her name was)).

The company i worked for was purely government sponsored and funded (we were writing educational software for special needs) and so we were governed by civil servants. Therefore we worked to deadlines and when i started i was very proudly told we never missed a deadline.

So it came to pass that the current project we were working on was coming to deadline and it wasn't what you would call finished nor was it gonna be and i expressed my concern to my manager and this is his direct quote

"On the night before completion day a very mystical thing happens ... all those bugs turn into features."

And you know what ... they did

[Tryagain]

They have already done more than the last owners in such a short time. I am pleased that anything is getting done. The communication is 1000% better than the last owners too. They are doing an A+ job in my mind. I applaude all they have done. Play the game and see that things are changing. I've been playing since beta and have seen most that went on. It is good to see things happening again. A couple of months ago I thought the game was done...it had no vital signs, but VI has saved it. I know you can't please everybody, but have patients please.

Diabolis
Order Shard

it wouldn't take much to do better than EI PME...

Vitrium ( Tulga V2) seems to be continuing off where Tulga left off.

[Vahrokh]

Well, the other posters have handled this pretty thoroughly, but i figured I'd just confirm it. There's a trick to giving project estimates when you're a programmer. 1) Estimate how much work you think the project is. 2) Now estimate how much work it would be if everything you can think of goes wrong. 3) Divide that into 8 hours days. 4) Double that, because you're lucky if you get 4 heads down hours of work done on a project a day. 5) Lie to your manager, because you know something unexpected will happen. 6) Pray that the things you didn't think of that go wrong balance out the things you did think of that didn't.

It works about half the time. If you have a good manager, he understands that an estimate is exactly that, an estimate. If you have a bad manager, you get called out on the carpet for going over projection

You have to be lucky. I worked for several software companies and they would never accept this.

Here, customers (corporations) continuously tell you how they could get an Indian programmer for 1/10 of the price. They make time capped projects (by people who have a shallow understanding about the matter) and impose the deadlines upon you. And wherever I go work for, the customers are all like that.

Finally, the commercial department tells "we could not ask for more than X days, competitors promised mountains and moon for that date and low price". Or the project is subject to be given to the competitor asking the least and shortest.
In any way there's almost always someone who decides time and it's never based on the programmers "needs". If they see a thing "looks like" taking 10 days, they sell it for 8 days (supposing always everything going for the best) and then press on that date "because the customer only pays for 8 days and if you won't finish in time we pay penalties".

During that 8 day project, naturally, while working overnight, a phone call arrives: "Mega important customer YYYY are halted by a problem, drop everything and run there".
You return 2 days later. "Welcome back, remember that now you have but 6 days left".
And these are all mission or even life critical projects, you can't deliver a piece of rubbish, have a worker cut in two by a badly programmed machinery and say it's "a feature".

Everywhere I go, every company I go work for, they are all, all like that. In fact I will stop programming as soon as I find anything different.

[Magfuddle]

Vitrium ( Tulga V2) seems to be continuing off where Tulga left off.

Having left in those days I am not sure its a good thing to say that is what I have come back to

[Vahrokh]

Ignore him, it's years he writes pessimistic like that.

[Kulamata]

Having left in those days I am not sure its a good thing to say that is what I have come back to

It would be worse if it were true.

[Arietna]

any type of development is like this. I work for a company that develops new technical instruments.. (I am the business manager, also website dev (a little bit) and database admin (a little bit))..

the one in particular I'm thinking of has so far taken 3.5 years in development and has been back to the "drawing board" stage at least 5 times when they found a technical problem they couldn't overcome.

THANKFULLY the guy in charge of instrument development also happens to be the owner of the business (believe me, he's a technical genius, just has way too much other stuff to do).. so there's no problem with this dragging on.

[Shirewood]

It came to me in a dream last night why it's taking so long to get an answer on the Unity characters. It's all about the exchange rate of the currency. They have to adjust the Mark, Euro and the pound to current exchange rate. As soon as they get it all sorted out.. the dollar loses value and they have to redo it.




ok.. I should probably have slept a bit longer.

[Menkure]

*Offers Shirewood some of Awdz's best freshly made Klava*

Here, this should wake you up!

-Menkure

[Spirit]

Is there anyway this thread can be renamed "Unity: An Update on Progress" so us peeps can see it?

Many thanks;

Rakku

[tramsan]

This ain't The Unity Thread, though. That one's burried back at page 3.....

[Spirit]

This ain't The Unity Thread, though. That one's burried back at page 3.....

lol.. even more reason to rename this current one and kill the old thread.

[Shirewood]

Actually, I started this thread just to ask what else they may be working on for Chaos/Order shards. I love reading about new things coming.. lol I have no patience. It's like new games that are being reviewed.. I gotta read the reviews.. and get grumpy when there aren't any. Like all the rumors about a Diablo 3. I played the heck out of Diablo 2.. Would love to see a Diablo 3.

I guess I am just a little over excited about playing the game again.



Though... I am anxious to hear about what is in store for you Unity players as well. That will be an awesome day when you folks are exploring and taming the world of Istaria again on your server or another.

[Kulamata]

My guesses as to what's being worked on:

1. Vista
2. Unity transfer
3. Client performance improvements
4. Server side improvements so that the spawn system can be improved. (Number one in the poll of desired improvements.)
5. An Event. Winter Solstice perhaps.
6. Confectioner of course.

That'll keep 'em busy for a few...

[Goriax]

Here, customers (corporations) continuously tell you how they could get an Indian programmer for 1/10 of the price. They make time capped projects (by people who have a shallow understanding about the matter) and impose the deadlines upon you. And wherever I go work for, the customers are all like that.

Heh, you struck a chord. Billions of dollars have been spent trying to find a magic bullet which would make software development as predictable as building houses. And it did no good. But you can find a LOT of crap on the market about all sorts of approaches which are supposed to do wonders, except they never seem to work out. So they're replaced by more crap.

I think all companies like to be as you describe. Many will play all sorts of head games. But if they REALLY thought they could get an Indian programmer to do it for 1/10 the price they'd have already done it. They wouldn't be talking to you. My experience doesn't sound quite as bad as yours though.

What I've observed is that companies that act like that usually end up with projects which are horribly late, buggy, and missing features. Because they just didn't spend the time to do the job right.

The last company I was at was typical. For some reason they decided that the marketing department should make all the decisions. Mangers in other divisions who didn't agree with whatever crap marketing spewed out didn't stay managers for long. So naturally everyone went along.

I recall the first project I worked on there. This was to be a shrink wrapped desktop data mining application to be sold for about $1500. When I showed up for work they said it was due to be released in six months and virtually no work had been done on it. They were still deciding on screen designs (which went on for another two months -- marketing was involved). Three months into the project I saw a press release describing all the great features that were to be in this product. I was supposedly building this thing and hadn't heard of half the features. The bottom line is we worked hard, it was released three months after the release date the marketing department had decided on and was missing many of the features the marketing department had decided were going to be in it. Marketing was not pleased. In their minds it was of course the fault of development that their timeline could not be adhered to. Never mind that any developer could have told them what was going to happen if they'd let people speak up without fear (I did tell my boss repeatedly what was going to happen but of course she wasn't going to jeapordize her job by saying so to her superior).

It wasn't until version 2, when they finally got the glimmer of a clue, that we were able to put out a decent product. That initial messed up release, plus other fubard projects by the same mismanagement caused the company's reputation to take a nosedive. Heh, even one of the founders (who was running another company at the time) was on the verge of going to a competitor.

Oh yes, they lost so many of their developers that the head of development was practically begging people to stay.

In short, their refusal to deal with reality destroyed the company.

After that I could really appreciate the Dilbert comics when he portrays marketing as living in a fantasy world.

If a company decides on a development timeline without talking to any experienced developers then they're playing in fantasy land. If you own stock in that company you probably want to sell it.

I've been at my current company for six years now and intend to stay longer. It has a remarkably low turnover rate among development staff. Probably because management really does understand development and is willing to meet people halfway. And they don't play head games. With few exceptions they've made more money every year I've been here. Yes, such companies exist, but you'll have a harder time getting a position in one because of their low turnover -- they don't have to be in constant recruiting mode.