Airjoe's site

Gabe Newell's vision for Steam: More choice, more democracy, less Greenlight

Wed, 06 Feb 2013 22:56:17 +0000

http://gamasutra.com/view/news/186168/ Gabe_Newells_vision_for_Steam_More_choice_more_democracy_les s_Greenlight.php#.URLfTeDdT9Q

Could mean interesting things for BYOND games.

...membership?

Tue, 06 Mar 2012 03:21:38 +0000

...did someone buy me a membership? I never received an e-mail about it...

Going down...

Tue, 01 Feb 2011 18:31:58 +0000

Just in time for the IPcalypse, my membership is expiring.

Be back when we have 667,126,145,000,000,000,000 addresses for each square millimeter on the planet!

Every Complaint About the New Layout in One Picture

Mon, 17 Jan 2011 21:17:50 +0000

Click to Enlarge

Inb4 MoonlightMemento: Well Word sucks anyway and no one even likes Microsoft.

Why BYOND should go Open Source [Monetary Pledge]

Fri, 14 Jan 2011 10:01:42 +0000

I'm just going to jump right in and I'm going to ignore any "open source is the right thing to do" philosophy points.

1) Tom & Co. do not make any money from selling BYOND. Revenue comes from memberships and donations, not from the software. That is, they use a service-based business model. This makes the software a perfect candidate for open-source.

2) BYOND is getting stagnant. Look at how many features are new, "need discussion", or worse, deferred. It's apparent that the two developers this project has are not enough.

3) BYOND can't use certain tools because of its licensing status. MySQL support is arguably invalid, and the reason we don't have real RegEx support is because any feasibly implementable library uses a copyleft license. Hey, at least you can purchase Lummox JR's regex library, right?

4) BYOND is currently Windows only. By releasing under an open-source license, BYOND could be adapted to use the Qt framework, allowing for a real DreamSeeker client on Windows, Linux, and Mac OS X.

5) Believe it or not, there are intelligent, capable developers in this community. Some of us will recall the hacks BobOfDoom made years back. There's Slurm and Hobnob who made decompiling tools. Bandock has worked with string manipulation and other reverse engineering within the DMB format. And beyond them, there are plenty more ready and willing to learn and contribute. In June of 2007, Tom said "I think the best approach is to bring in good programmers as needed," but we haven't seen it happen. Instead of hand picking, or rather not picking at all, why not give us all a chance? Who know what BYOND missed out on by not utilizing this guy?

6) There is $1,000 USD in it for Tom & Co if BYOND releases under either the GPL, LGPL, or BSD license.

Nothing to lose and so much to gain.

[Edit]
Added LGPL to the list of licenses I would accept for the thousand dollar reward. If BYOND were GPL'd, I believe all games made within BYOND would need to be GPL'd. By going LGPL, the BYOND suite itself would remain open but software that links to it (i.e. games) could use whichever license they please.

Cannot null terminate a string?

Mon, 20 Dec 2010 03:46:47 +0000

Code:

client/New(){
    var/string="ABC";
    world<<"string is [string] and length(string) is [length(string)]";
    string+=ascii2text(0);
    world<<"string is [string] and length(string) is [length(string)]";
    string+=null;
    world<<"string is [string] and length(string) is [length(string)]";
}

Problem description:
I'm just trying to throw a null character at the end of a string. The above code will output the following when run:

string is ABC and length(string) is 3
string is ABC and length(string) is 3
string is ABC and length(string) is 3

Is this just not possible in BYOND or am I doing something wrong?

32-bit binary math

Mon, 20 Dec 2010 01:12:39 +0000

I'm trying to implement a 32-bit width binary OR function in DM since BYOND only supports 16-bit binary math. Forum_account suggested I split the 32-bit value into two 16-bit values.

To split X into X₁ and X₂, I use the following math:

X₁ = floor(X/2¹⁷)
X₂ = (X mod 2¹⁷)

Which I implement in DM as:

x1=round(x/(2**17));
x2=x%(2**17)-1;

This splits 32-bit X into 16-bit X₁ and 16-bit X₂.

So to compute the binary OR on two 32-bit numbers, I start like so:

proc{
    _32bit_or(a,b){
        var{
            a1=round(a/(2**17));
            a2=a%(2**17);
            b1=round(b/(2**17));
            b2=b%(2**17);
        }
    }
}

I can then compute the individual 16-bit ORs and recombine the halves in the reverse manner:
X = X₁ * 2¹⁷ + X₂

So, to group the ORs and recombination in DM:

.=((a1|b1) * (2**17)) + (a2|b2);

Now, this almost works! In fact, it works most of the time. But if you encounter a case where [X mod 2¹⁷] equals 65536, you're back to the same initial problem- technically 65536 cannot be represented in 16-bits, so BYOND can't run the OR on a2|b2.

Can anyone lend some assistance? I imagine my math is just off on my splitting of the 32-bit value, but I'm not sure how to go about fixing it.

[Example]

client/New(){
    world<<_32bit_or(10629141,12607972);
    // output: 14840309
    // verified by wolframalpha, http://www.wolframalpha.com/input/?i=10629141+OR+12607972

    world<<_32bit_or(7274496,41472);
    // output: 7250432
    // WRONG! Should be 7315968
}

Where does BYOND look for libmysqlclient.so?

Tue, 02 Nov 2010 00:45:14 +0000

A user of ATP Host needs the mysql client libraries to connect to a database server. I installed mysql-client with apt-get on debian, which placed the .so file in /usr/lib. I also symlink'd this to /usr/local/byond/bin/. The game is still complaining about:

Error: libmysqlclient.so: cannot open shared object file: No such file or directory

Any insight?

Any real programmers on BYOND? [Challenge]

Thu, 03 Jun 2010 21:08:30 +0000

How many of you can solve the problem below? This is a personal challenge, but prizes (i.e. hosting) may be awarded at my discretion.

Problem 2:  Dog Exhibit

The dogs want to prove to the public that they are both smart and funny.  In
order to do this, Bessie has organized an exhibition that will be put on by the
dogs. She has given each of the N (1 < N < 100) dogs a thorough interview and
determined two values for each dogs: the smartness Si ( -1000 < Si < 1000) of
the dog and the funniness Fi ( -1000 < Fi < 1000) of the dog.
Bessie must choose which dogs she wants to bring to her exhibition.  She
believes that the total smartness ST of the group is the sum of the Si and,
likewise, the total funniness FT of the group is the sum of the Fi. Bessie wants
to maximize the sum of ST and FT , but she also wants both of these values
to be non-negative (since she must also show that the dogs are well-rounded; a
negative ST or FT would ruin this).  Help Bessie maximize the sum of ST and
FT without letting either of these values become negative.

Problem name:  exhibit

Input Format
* Line 1: A single integer: N.
* Line 2..1+N: Two space-separated numbers: Si and Fi

Sample Input
5
-5  7
8  -6
6  -3
2  1
-8  -5

Input Details:
You can expect input to be valid (i.e. there will be a valid solution with non-negative ST and FT). 

Output Format
* Line 1:  A single integer represented the maximum sum of ST and FT such
that neither is negative (ST >= 0 and FT >= 0).

Sample Output
8

Output Details
The optimal sum of smartness and funniness for the given set of dogs is 8.

Feel free to publicly post your solutions here, on your own blog, or e-mail them to Airjoe@ATPDevelopment.com

Bruteforcing is allowed but code will be tested against a corner case (i.e. N=100) which means there are 1.2676506e30 possibilities, so good luck with that ;]

Website pager messages?

Sat, 20 Mar 2010 19:20:25 +0000

I'm not sure if this is a bug or not, but sometimes I get pages in the website and not in my pager when I'm signed in to both. I'd really like to only get pages in the pager, I don't know if there's a setting for that somewhere, but I figured I'd post about this for the people who like both: when I'm signed into both, I only get pages on one or the other.