OSDev.org

Welcome.
It's been ages since I decided to provide an update here, so I thought I'd give a progress report.

Here are some of DiNS current features:
[*] Intel Xeon optimized
[*] 64-bit (Long mode) support
[*] SMP support (up to 16 cpu's)
[*] Paging 1:1
[*] HDD support (LBA28/48)
[*] Arbitrary-precision integer support
[*] ARP/RARP/ICMPv4+6/DHCP/DNS support
[*] Minimal HTTP server
[*] Logging support
[*] 30+ P2P/BitTorrent client filters

DiNS can be installed and booted from a hdd now and has proven to be quite stable. I added a lot of error catching/fixing to the hdd driver code which drastically improved things .

The web-based administration site has been going great. It can be logged into and the admin can view live stats such as the amount of packets in/out per card, uptime, cpu-count, p2p packets filtered, and will soon have the ability to dip into the logs saved on the HDD which contain more specific info about dropped packets (type, reason, timestamps, eth card, etc...).

I have been having regular meetings with my attorney and various investors and consultants lately. I currently have been invested in twice and we are working towards and third very large investment that will set me on a year-long plan to further develop and prototype DiNS into a market-worthy device with the aid of a few developers and tech-savvy consultants/aids.

I gave a brief demonstration recently and was awarded some support to the project by well-established and reputable businessmen who could definitely come in handy later. At the moment we are scouting out potential test sites during alpha and beta stages.

So all-in-all, things seem to be looking up for the DiNS OS project.

Thanks for tuning in.

Nifty!

Any chances of seeing screenshots or a video?

And BTW, it still says v2.26 in your signature.

Screenshots.... hmm. I suppose I could get around to getting some stills. I'll get back to you on that one.

Yeah, I should probably change the version and the current work bit as well.

I'm not sure about the video at this time though, I'll see what my attorney thinks and if he gives it an OK then I'll gladly post them.

All right, sounds good!

I assume you're planning to sell this OS?

I assume you're planning to sell this OS?

Yes. I intend on selling to corporate and educational facilities that could use file-sharing blocking.

that's awesome, it's certainly gone a long way since i was last on this forum.

i'm glad it's going well for you, it's about time we had some variation in the commercial OS market

Hey

Good to see some updates. I see you worked really hard and far as i know you are of the the biggest and only closed source operating system. I take my hat off to. As Tony said something of screenshots i would like to see as well if you allow. By the way what is your website address?

Patrick

Hi,
Well done .Congrats from an idiot .

Regards
Shrek

Shrek wrote:Hi,
Well done .Congrats from an idiot .

That is a "good one" because only an "idiot" could believe that this is a good thing.

Let me be the one that stops his dreams and makes the reality wake up call.
I do this from kindness because otherwise "01000101" will waste his time living in high hopes and dreams.

Of course "if" this DINS dream is true because we have not seen any binary releases or at least any kind of screenshots and hence all of this could be just "talks" or "dreams" at a more deeper level.

However let us assume that his project is fair and true and explain WHY it stands NO CHANCE of becoming rich or famous by doing this.... well, unless he already is rich or his daddy wants to pump up a few billions or euro into his kid's dreams.

I have on my desk a device that would kill DINS at first sight.

It is an embedded router made by a huge corporation. It has a few (4-8) Ethernet ports. It acts as a switch for LAN and as a gateway / router for WAN. It has a "hardware" firewall inside. It has a DNS server and DHCP server. It can act as a mini web server. It has throttle control.

And YES it has packet filtering and P2P network filtering. It knows about 2000 P2P networks and games and applications and ...you name it ... and it can be updated on line. And of course you manage them by HTML and by telnet and by RS232 and by USB and encrypted if you like to.

And it all costs about 10 euros. There are devices that cost less or devices that cost a lot more (more network connections and more features) but the decent ones are pretty cheap because they are mass produced.

There are approx 10 free OSes designed especially for this kind of task (network firewall, P2P filter, router, gateway). They are ALL free and open source written in C and have a lot of features. Hence the OS/software in this area values NOTHING. The only real problem is interfacing with custom hardware and real life testing against all kind of exploits.

What is important is how you adapt it to the hardware. Do I need to say that NONE of those hardware devices uses an x86 architecture? Remember: NONE is x86. They are all custom CPU and custom chipsets or embedded architectures that are very different from x86.

Also none of them use the "normal" Realtek or NE2000 network chips. Even if they are produced by Broadcom or Realtek they are designed specifically for this kind of task not for x86 interfaces like PCI etc.

Hence x86 kind of drivers value nothing in this world.

They use custom specific chips that are made on purpose for this task. The specifications for those network and CPU's and chipsets are "free" IF you pay a lot in the range of 10.000 - 1.000.000 euros at least. And you must buy 100.000 pcs of them for "a start".

None of this is x64 and hence adding x64 to such an appliance is "childish".

Nobody would buy a costly x86 computer just in order to drop in a "dubious security" when they can buy a certified appliance from a well known corporation (think IXIA or Thompson or Zyxel or Cisco here) that has 1000 programmers continuously adding new P2P filters and doing unit testing and regression testing and offering a safety in case of a breach.

Corporations can produce millions of devices at a very low cost... and they can switch CPU's and chips and fabric all over the world and have 1000 programmers at their command ... can you?

If network admins want to use a x86 architecture then they would choose a mini Linux distribution that would make good use of that x86 CPU power and at least the network administrators are comfortable with iptables firewall and well known security programs on Linux/ Unix.

Network admins are usually also GPL or Windows Server freaks and they would NOT use a closed source or unknown OS and they do advise or sometimes decide what kind of firewall or network security a company will buy and use.

Hence you waste your time....

You should drop those dreams of rich and famous and try to think with no desires. This way you will see and think more clearly.

Choose something that suits your skills and make you feel JOY (not money ) and DO NOT strife for rich or famous because that is a waste.

Rich people came and go and none of them make any difference in existence and they are simple ignored because the struggle and strife fills them with darkness inside while they smile and hope on the outside.

If you stay in JOY area THEN "may it be" that One day you will notice that you are truly "rich inside"...

Otherwise you might eventually become rich in the outside world but inside you will notice that you are "still a poor child dreamer"

This criticism is supposed to be taken as "a pointer" my friend.

Going in here for 01000101's defense.

bontanu wrote:Of course "if" this DINS dream is true because we have not seen any binary releases or at least any kind of screenshots and hence all of this could be just "talks" or "dreams" at a more deeper level.

That's because it's closed source, and he's been telling us that it has no interface whatsoever!

I have on my desk a device that would kill DINS at first sight.

Your device works as a hobbyist OS for a large-scale server or gateway? Nifty, post some screenshots.

It is an embedded router made by a huge corporation. It has a few (4-8) Ethernet ports. It acts as a switch for LAN and as a gateway / router for WAN. It has a "hardware" firewall inside. It has a DNS server and DHCP server. It can act as a mini web server. It has throttle control.

Most people can buy those for a nice little sum of cash. Next.

And YES it has packet filtering and P2P network filtering. It knows about 2000 P2P networks and games and applications and ...you name it ... and it can be updated on line. And of course you manage them by HTML and by telnet and by RS232 and by USB and encrypted if you like to.

TLDR.

There are approx 10 free OSes designed especially for this kind of task (network firewall, P2P filter, router, gateway). They are ALL free and open source written in C and have a lot of features. Hence the OS/software in this area values NOTHING. The only real problem is interfacing with custom hardware and real life testing against all kind of exploits.

But they're not actively developed by a single, solitary hobbyist.

Didn't read the rest but here's a quick summary: Mr. E's OS runs on standard hardware, so you can turn an old laptop or 1990's server into a better network box. I'd prefer to pay for that, knowing that 01000101 wrote it and I could fix the hardware when I needed to, than shelling out the cash for a proprietary box I know nothing about.

TLDR version: I call the post I'm quoting arrogant BS.

Troy Martin wrote:Didn't read the rest but here's a quick summary: Mr. E's OS runs on standard hardware, so you can turn an old laptop or 1990's server into a better network box. I'd prefer to pay for that, knowing that 01000101 wrote it and I could fix the hardware when I needed to, than shelling out the cash for a proprietary box I know nothing about.

You have to remember that corporations work a little differently than technical people at home. Even if they did want to use "repairable" old hardware, I have a hard time understanding why they wouldn't just use Linux, or something similar. I say this thinking of the (stoic, conservative, just plain scary) network managers I've worked with.

A question along those lines (the first one I'd ask if being approached to purchase) is: "Why didn't you just add the features onto Linux or BSD?!?" And this question is really about why I should trust the developer(s) judgement. While I love re-inventing the wheel (out of curiousity) at home I avoid it like the plague at work. If I were E, or his salespeople, I'd have a rock-solid explaination prepared for this question. Because they'll be entrusting their "precious" network to you.

I actually used to work for a small network security company. What I learned from them, if anything, is that the easiest way to sell a widget is to become a network consultant, then at some point advise a client to buy your product. It is the best way to avoid technical people.

Troy Martin wrote:But they're not actively developed by a single, solitary hobbyist.

Could you expand this statement? I'm curious what you meant (and I've interpreted it in conflicting ways). Despite different interpretations, the "What happens if that person gets hit by a bus?" question is omnipresent.

For the second quote, I was saying that 01000101's doing the project by himself, not in a group with other programmers.

Troy Martin wrote: Going in here for 01000101's defense.

I guess he is able to defend himself.

That's because it's closed source, and he's been telling us that it has no interface whatsoever!

Well, I did say that I assume his is true.

However can I also have the right to express some doubts after such a long series of "talks" about a "ghost" OS and getting rich and lawyers and his investors?.

Esp when other people that make such claims with no screen shots or binary to test are kind of ignored here or openly challenged (and with good reasons).

Your device works as a hobbyist OS for a large-scale server or gateway? Nifty, post some screenshots.

Here you have your screenshots:
http://www.oby.ro/files/PC130007_b.JPG
http://www.oby.ro/files/PC130009.JPG

The first one is programmable by a hobbyist in many ways.

The second one is the one with 4 Ethernet ports firewall and P2P network filtering and can be programmed in C eventually and it does have a lot of interfaces including command line and http and telnet.

They do rest on my new glass tables

I have others but I guess you can find some your self in the network admin area of your company or at the company where you work.

Most people can buy those for a nice little sum of cash. Next.

That is exactly my point.

Most people will buy those and will not buy his OS. Mainly because his hardware will cost more and perform less for the same money.

I do not challenge the making of his OS as a hobby. I challenge the "getting rich and famous" part and the economical validity of his approach.

I do it not to "arrogantly BS" him but as an kind advice in order to spare him and his investors a lot of loss in money. It might not be late to change his targets with the very same OS code he has done until now and find a better business model.

TLDR.

Interesting term.

But they're not actively developed by a single, solitary hobbyist.

That is not a good selling point. Not to a corporation or a big company. Or even to a small company. They do not like to do business with or to be dependent on a single person with limited development caps and assets

Who will run the customer answering lines?

Didn't read the rest but here's a quick summary: Mr. E's OS runs on standard hardware, so you can turn an old laptop or 1990's server into a better network box.

I know that. But do you really think that a corporation or a company still has an "older PC" availble to turn into a new thing and they want to risk using it?

Then you are dreaming and do not know how things work inside a company or a large corporation. They will prefer buying some new hardware device 99 out of 100 times.

I'd prefer to pay for that, knowing that 01000101 wrote it and I could fix the hardware when I needed to, than shelling out the cash for a proprietary box I know nothing about.

Nice to hear that you want to buy it. But surely his business model doe not base of selling to a single person? The cost would be huge in this case.

How can you fix an old box? Do you still have ATA IDE arround in SATA times? do you still have a PS/2 mouse? a PS/2 keyboard? a floppy disk unit? For the next 5 years?

And a proprietary brand name box does come with a warranty.

I doubt the old PC box still has a warranty and a company cares for this. The warranty can get it switched with a new one in no time if the hardware breaks.

But you know what? Those embedded stuff rarely break. They simple become obsolete...and at the cost of 10-100 euros... you can buy a new one any time you need one...with a new warranty.

And I thought he intends to build his own hardware. Why did he add x64 features to his OS then? he expects an old PC to have x64 and multiple CPU's ?

Anyway big company are very reluctant on buying new software unless it is Windows, or Office or Visual studio or Photoshop/Autocad or antivirus/firewall or custom accounting (SCALA) coming from a big well known and proven company.

If they have an "old" x64 PC then they will eventually use Linux/Open BSD on it.

TLDR version: I call the post I'm quoting arrogant BS.

Thanks for explaining the term.

I was just stating the truth as I know it as somebody that has worked in and with corporations and big companies and I own my own company and I know the politics for new acquisitions.

It might look like logical to you to use old hardware plus a hobby OS but a company wants new hardware and to work with another big established company that has the caps to service them rapidly all over the world and replace or fix things fast and they need a warranty as mandatory politics esp for a critical sub system as the network security.

My advice might look like arrogant to you. In this case please excuse.

bontanu, have you ever read Paul Graham's site?

A startup with a couple founders in their early twenties can have expenses so low that they could be profitable on as little as $2000 per month. That's negligible as corporate revenues go, but the effect on your morale and your bargaining position is anything but. At YC we use the phrase "ramen profitable" to describe the situation where you're making just enough to pay your living expenses. Once you cross into ramen profitable, everything changes. You may still need investment to make it big, but you don't need it this month.

You can't plan when you start a startup how long it will take to become profitable. But if you find yourself in a position where a little more effort expended on sales would carry you over the threshold of ramen profitable, do it.

Investors like it when you're ramen profitable. It shows you've thought about making money, instead of just working on amusing technical problems; it shows you have the discipline to keep your expenses low; but above all, it means you don't need them.

There is nothing investors like more than a startup that seems like it's going to succeed even without them.

stephenj wrote:bontanu, have you ever read Paul Graham's site?

No, but I did read it now. I kind of live by that standards.
It also say "Do not take rejection personally"

So if we go economics, how will he make it "ramen" profitable?

2000 Euro per month means selling an equivalent of 200 OSes per month at 10 euro per OS in order to compete with embedded mass produced devices.

I have seen much more experienced teams of 2-3 people with extraordinary good GFX and games products not selling that much software in a month.

Not mentioning IF he wants to build the hardware himself and the costs of that hardware. And that can eventually mean 200 calls to the call center in one month if the customers have to install it them self.

And why would an admin that goes shopping for the next device choose a software only solution that needs installing on old hardware instead of hand picking a hardware box at the same price that only needs to be plugged in, no installation, just a little setup in a html page and then you can forget about it for the next 5 years?

And with 2000 euros per month he can not make his own hardware cheaper than custom ASICS made by Broadcom Corporation and his software will not have better testing and hardware support than the one made by IXIA corporation

New startups survive like that in a market share/target that is all new and not yet filled up by a few corporations that extract all profit that can be made or if they offer the same as a corporation but at a much lower price like BeoS did with a standard GUI OS at 20 euro compared with Windows are 100 euros. However if you compete a corporation then you should expect harsh retaliation.

The problem here is that you can not go lower that those custom made hardware sold today for this job. Not without a lot of cash and costly aggressive marketing.