How hard would it be to do these following things

iggyman

i was just lookin at some threads that people have posted in about there i.t jobs...sayin they hate it..because of that..i was thinkin about u's startin your own business...im asking yous now how hard would it be and what is involved to create the following programs with your own new ideas that bring new stuff to the tables of the current programs out today..

1>a media player
2>a virus protection program
3>a high-end pc game
4>an internet browser
5>and my most curious one an operating system

please reply with your toughts on this..

aidan_walsh

Simplistic, to say the least, but anyway.

First you would need ideas. Those are always a good starting point.

Next you will need skill with a language of choice, and the various principles of software engineering. This takes a few years of work to become comfortable when starting from scratch.

Knowledge in the field of what you have decided you want to improve is a must as well. No point in trying to be innovative if someone already did it back in '73.

Then you will need to have the ability to turn those ideas into plans.

And last but not least, recognise that you have about a snowballs chance in hell of doing it all on your own. You will need to work with a team.

parasite

i, for one, look forward to iggymanOS !!!1

iggyman

thanks...ill have one dont worry...

rsynnott

There are two dominant varieties of operating system today. One is UNIX, from Bell Labs the 60s. The other is Windows, from Microsoft and IBM in the 80s. New operating systems tend not to catch on.

Hobbes

rsynnott wrote:

There are two dominant varieties of operating system today. One is UNIX, from Bell Labs the 60s. The other is Windows, from Microsoft and IBM in the 80s. New operating systems tend not to catch on.

Wow Steve Jobs is going to be pissed. Come to think of it what about AIX?

OP.. It depends on what kind of program you are doing and how quickly to market you want to get it.

Of those a Media player would probably be the easiest as you could plug in existing codecs.

The rest would require a team of people.
- Manager
- Architect
- UI developer
- R&D
- A few monkey coders

Probably more.

Mutant_Fruit

5>and my most curious one an operating system

.
Assuming you want a versatile OS not based on linux, at least 100 developers and a few years of your life.

Unless you're going for specality OS's such as what drives an MP3 player etc. You could probably knock one of those up yourself in a few months (depending on how much time you have to spend on it and how good you are).

rsynnott

Hobbes wrote:

Wow Steve Jobs is going to be pissed. Come to think of it what about AIX?
.

MacOS X and AIX are both UNIXy things. MacOS pre-10 was not. It's gone, though, more or less.

Merrion

The first thing to do if you want to start an IT company is find a niche.
Since all of the products you mention are niches that have already been (over) filled you won't make any money from them.

Why not instead make a list of things you would like to have (in terms of programs) but which aren't already available - then see what is involved in coding them.

fatherdougalmag

I think the place to be looking in is the fusion between different areas and software. Like applying user-end software to existing appliances thereby giving them intelligence. For example, look at how TV menus, printers, etc. have been granted a new lease of life by adding some minimal user interface and extra functionality to them. Satnav in cars is fairly common now and this will probably evolve further.

So, look around you at everyday boring objects and think about how a bit of software could add to it's functionality. There's oodles of things that you could come up with!

bonkey

fatherdougalmag wrote:

So, look around you at everyday boring objects and think about how a bit of software could add to it's functionality. There's oodles of things that you could come up with!

It should be pointed out that >99%* of the results which come out of such thinking are typically what I call "solutions looking for problems" rather then "solutions to problems".

I've lost track of the number of "smart fridge" ideas I've heard. TV built in, with RFID tags noting what goes in and out, and automatic re-ordering/reminders/whatever......

I just don't get it. I know whats in my fridge. I know roughly how old things are, and when they need to be used by. When something runs low, the person using it adds it to the shopping list. I have no need for a TV built into the fridge for radio, tv, internet access, or anything else. And just like I don't buy TVs with built-in DVD players (failure in one == replace both), I will keep my radio for listening to the radio, my tv for watching tv, my cookbooks (and internet + printer) for recipes and my fridge for fridging stuff.

Despite what it may seem, successful ideas aren't that common. This is ultimately one real reason why so many of those bottom-line-focussed managers exist that everyone has been complaining about. They exist because there are a huge number of companies who won't stay in business if they don't keep their margins tight.

If you have a great idea to fill a niche, or can produce a genuinely better mouse-trap, or you get lucky, you'll get rich. Otherwise, you'll focus on that bottom line and struggle....and your employees will complain about you and your managers just as they have here about others.

jc

*I made this number up

fatlog

as has been mentioned the suggested topics are simplistic and completely over done.

in my opinion you should be looking at trends and where technology is going. it has been dismissed in a previous post but things like electrical commodities will soon start to get smarter. there will be a push at some point to completely integrate the home kitchen. how successful it will be is anyones guess but it will have its day like most things. technologies like auto-id are in my opinion worth knowing.

other things to look out for are projects that are in the pipeline. For example, the European Space Agency has started to build and deploy satellties for a European Satellite Navigation System called Galileo. Its based on something like 30 satellites and will cover all of europe when it is up and running, which they estimate to be between 2010 and 2015. they estimate it will create around 100,000 jobs for companies who want to develop tools that use the array of satellites. They built it with this in mind.

i'm not saying any of these are easy or trivial. but they are emerging technologies and they are the places where you should really be looking if you want to develop something new.

SolarNexus

1>a media player
2>a virus protection program
3>a high-end pc game
4>an internet browser
5>and my most curious one an operating system

1). easy enough. Using windows API's, it can be as simple as just creating a graphic interface over what is essentially windows media player. Or if you were to create from the ground up... might take you a good long time, especially considering supporting all the filetypes etc.

2). I would expect its extremely complex, though I honestly dont know much about it, I am sure such things as heuristics algorithms take a good long while with a lot of work

3) are you talking about making one yourself? dont even bother, seriously, I tried. So unbelievebly involved you would not believe it. You need to create the models, the graphics, the interface, the networking components, the 3D engine, the physics engline, the 'scoring system'. There are so many elements to a game, it takes a team of 30+ over 2yrs these days.

4) if you want to use pre-existing apis or open source stuff, its simple as pie, just get an interface, whack it on the decryption engine and your done. If you want to build from ground up, its going to take you ages. With all the web standards you have to adhere to, its not worth the effort starting from scratch but better to 'fix' pre-existing open source. Even that would take ages.

5) Oh My God. I'm serious. You have no idea how complex this is; it took the open source movement - what? 10 years to come up with the goods. Its just far too complex. I dont know where to start... no chance in hell, basically. its like trying to fly to the moon with spoons.

Its like everything, if you have enough $$$ and enough man power you can do almost anything. Sure, each one of those could be done, relatively without much hassle, given enough time and enough money. But thats the problem, software development is like any other buisness, its got to be now and its got to cost next to nothing while making us a fortune.

I'm not saying you cant make software yourself, its just you chose the most involved, most complex options out there, most of which have been in progress for years now.

Heres a list of realistic software development options:

• FTP client
• Email Client
• Instant Messanger
• File encryptor / decryptor

Like most things these days, its best in a team of like minded individuals, with the skills and abilities to aquire a goal whilst having the wish and determination to reach for that goal.

rsynnott

SolarNexus wrote:

5) Oh My God. I'm serious. You have no idea how complex this is; it took the open source movement - what? 10 years to come up with the goods. Its just far too complex. I dont know where to start... no chance in hell, basically. its like trying to fly to the moon with spoons.

Actually, writing a simple operating system is easy. Writing an operating system that people want to use is more difficult. The only attempt in the last decade that had any success was BeOS.

Mutant_Fruit

Replace "BEOS" with "Linux derivitives, and my personal choice, BEOS." and you'll be a lot more accurate

In reality, consumer OS market is never going to change from Wintel.

bonkey

Mutant_Fruit wrote:

In reality, consumer OS market is never going to change from Wintel.

And in the 70s, no-one was ever gonna change from IBM.

Go figure.

In terms of starting up a company to develop something...you need a hole to fill. I see two types of hole here :

Hole 1) Building a better mousetrap:

The iPod is a great example here. Creative Jukebox had been out well before it, but the iPod just hit the right note. It did things right.

If you look at the list presented at the top...which one of those can you do better than whats out there? What cost and how much can you sell it for only become relevant after you can offer something better. (OK - you can also take the "just as good but cheaper" option, but seeing as there are free implementations of every item on that list, I can't see how you can undercut on price).

Hole 2) Building something that simply isn't there at present:

Example - have you any idea how annoying it is that my address-book(s) don't move around as easily as (say) my IMAP mailbox does? Sure...USB stick is a nice idea....except when I'm working on customer sites where I can't plug it in, or when I don't have it with me. And sure, I could hack some sorta web solution. But thats not what I want...I want it integrated into my mail client and be as seamless as IMAP is for mail in general. I'm also pretty sure I'm not alone on that one.

Hobbes

bonkey wrote:

The iPod is a great example here. Creative Jukebox had been out well before it, but the iPod just hit the right note. It did things right.

Both more or less the same. I would say it was more to do with snazzy marketing and brand awareness then anything else.

rsynnott

Mutant_Fruit wrote:

Replace "BEOS" with "Linux derivitives, and my personal choice, BEOS." and you'll be a lot more accurate

In reality, consumer OS market is never going to change from Wintel.

No, I don't use BeOS. Linux is not a new operating system concept as such, it is simply a rehashed UNIX. Ditto for WinNT/2000/XP/Vista, except they're rehashed OS2/WinNT.

bonkey

WinNT is a rehashed WinNT !! ?? !!

rsynnott

bonkey wrote:

WinNT is a rehashed WinNT !! ?? !!

Well, NT3.5/4, the ones most people are familiar with, were WinNT3.1 (a horrible nightmare, joint-developed by IBM and Microsoft) rehashed.

Capt'n Midnight

bonkey wrote:

And in the 70s, no-one was ever gonna change from IBM.

Go figure.

They had no PC's back then.
New platforms used to be a spur for new OS's but linux etc. and even windows come in embedded varities so in a lot of case it's not worth the hassle when you can just choose one off the shelf.

BTW: what runs on mainframes and minis today ?

As for OS's IIRC someone got Basic on to a PIC - an interpreter in 512 Bytes !
Not much of an OS but it had I/O etc.

WinNT owes a lot to IBM and Digital VMS. There is a lot of unix like stuff under the bonnet. Try start run %windir%/system32/drivers/etc
(yeah - "etc" folder with forward slashes )


I'd ignore the antivirus program since you have to know about every possible exploit and then some..

ressem

SymbianOS?

I want it integrated into my mail client and be as seamless as IMAP is for mail in general. I'm also pretty sure I'm not alone on that one.

Definitely not.
How opposed are people here to twisting RFCs to accomplish this sort of thing?
E.g. Courier-Imap supports a Outbox extension to allow mail delivery without the client using SMTP directly.
An imap-like frontend to Imap server/ Calender/ Contacts /Outbox client services seems completely sensible to me, even if the backend is Ldap/ical/postfix, and not as efficient as dedicated services.

Hassle making sure that imap clients don't break, but that exists with all imap servers, rfc is claimed to be contradictory by some implementors anyways.
For the imap like UI, How about address book appearing as vCards within a folder, or pseudo mails which can be replied to.

rsynnott

Capt'n Midnight wrote:

BTW: what runs on mainframes and minis today ?

Most of the world's banking.

bonkey

Capt'n Midnight wrote:

They had no PC's back then.

Exactly.

And with the rise of functionality on your PS3, on your mobile phone, and whatever comes next, you still reckon its safe to say that there's no new hardware platforms coming along to dethrone the multi-purpose PC that we all love and hate?

New platforms used to be a spur for new OS's but linux etc. and even windows come in embedded varities so in a lot of case it's not worth the hassle when you can just choose one off the shelf.

Again, I point you at the PS3. Lets not forget Symbian on mobile phones. And thats whats already big and getting bigger. Lets not forget to consider where iPod-style devices will evolve, and the fact that the world is still waiting for a proper "TV-OS" for the fully integrated home entertainment future.

MS have today's consumer market sown up. Just like IBM had in the 70s.

BTW: what runs on mainframes and minis today ?

I don't think I've worked for a client in over a decade who didn't have at least one major system running on one or the other (typically the former though).

WinNT owes a lot to IBM and Digital VMS.

Heh.

VMS++ == WNT. And the same lead designer worked on both.

But in fairness, most computer hardware and software owes a lot to what came before it.

jc

bonkey

ressem wrote:

How about address book appearing as vCards within a folder, or pseudo mails which can be replied to.

Pseudo mails were exactly how I thought about doing it. That way, non-mobile-addressbook mail clients would simply see it as a mail. Issues, however, would include :

- Prevent ppl spoofing your address-book
- Prevent non-smart mail-clients from allowing you to delete your mailbook
- Oh sod it...you get the idea.

Anyway....I'm not too pushed about going into detail. I posted the idea up more it show the type of thing that is definitely a hole needing to be filled.

jc

bonkey

bonkey wrote:

Hole 2) Building something that simply isn't there at present:

Example - have you any idea how annoying it is that my address-book(s) don't move around as easily as (say) my IMAP mailbox does? Sure...USB stick is a nice idea....except when I'm working on customer sites where I can't plug it in, or when I don't have it with me. And sure, I could hack some sorta web solution. But thats not what I want...I want it integrated into my mail client and be as seamless as IMAP is for mail in general. I'm also pretty sure I'm not alone on that one.

Before one opens one's mouth, one should check it isn't actually there.

Damn I feel dumb....considering its been around for two years. And some nice bloke has written a compatible Extension for Thunderbird.

Yay!

Guess I'd better cancel that company I was forming

jc

gerryk

Why try to develop a new product from the ground up? Why not become expert in one, or, better yet, integrating a few together and sell the service instead of an actual product.