PHP5 in New Zealand
sigurdm...@gmail.com
many New Zealand webhosts do not support PHP5 (I think ICONZ was
mentioned as an example). I say this because we've been using PHP5-
only features for quite some time, and moved our architecture from
PHP4 to 5 to get its great new features (OO, stability, general new
wad of functions, etc) back in 2005. We run our own servers so we can
choose to keep up with the current stable architecture of Apache2,
MySQL 5, PHP 5, etc.
With all of these people talking about the "horrors" of PHP, I am
wondering if this is due to people being unable to use PHP5 or unaware
that it exists, and akin to people out there with HTML4 tag soup and
needing to move into this decade and be XHTML with semantic code and
external CSS... Bear in mind you can easily turn a HTML4 doc into
valid XHTML but that doesn't really give you anything; its when you
actually spend time "refactoring" the code and taking true advantage
of the new features and thinking that you then get great mileage.
I see that everything from Jooma through to Cake all support PHP4,
which is about as limiting as supporting Netscape 4 was a few years
ago.
--
BTW, I'm only rushing into mailinglist with a few seconds of time, but
but I didn't notice a recent plug for the Wellington PHP meetup
tonight. Its being put on at Catalyst in Wellington, is free (or $5 if
you want pizza), and open to anyone interested in PHP, and we'll be
talking about PHP6 and imagemagick... Details/RSVP at http://php.meetup.com/346/
Sid Bachtiar
> which is about as limiting as supporting Netscape 4 was a few years
> ago.
Not everything, PHP Framework Symfony is PHP 5 only, and so is Prado.
Sam Minnee
SilverStripe is the __get, __set and __call functions to create all
kinds of Rails-like "magic" in the ORM :-)
Most people who I've spoken to about the issue are *able* to find PHP5
hosting with their provider or easily switch to a new one, it's just a
case of
One caveat is that some popular applications out there don't work in
PHP5 - only PHP4. This, frankly, is criminal because it's pretty easy
to fix (It took me about a day to upgrade the 50,000 line codebase of
SilverStripe 1) and it's holding the entire PHP development community
back.
What's going to happen when PHP6 comes out and we want some fscking
namespaces? Will we need to wait until 2015 before hosts actually
support it?
On Mar 7, 10:15 am, "sig...@silverstripe.com"
Nigel McNie
> With all of these people talking about the "horrors" of PHP, I am
> wondering if this is due to people being unable to use PHP5 or unaware
> that it exists, and akin to people out there with HTML4 tag soup and
> needing to move into this decade and be XHTML with semantic code and
> external CSS... Bear in mind you can easily turn a HTML4 doc into
> valid XHTML but that doesn't really give you anything; its when you
> actually spend time "refactoring" the code and taking true advantage
> of the new features and thinking that you then get great mileage.
PHP5 is still horrible. PHP4 is worse, PHP5 is still bad. I could list
all day about why it's a useless, poorly architectured language written
by a bunch of idiots who can't take criticism and have Zends finger up
their ass, but I'd be wasting breath I would say.
And I write a shittin of PHP every day, including a library in use all
over the world in thousands of blogs, wikis and forums.
Okay, so maybe many people who had contact with PHP4 now stay away
because they think php sucks. But showing them PHP5 is _not_ going to
bring them back. I know, I've tried getting people into PHP that way. I
don't seriously think that even showing them PHP6 will bring them back,
although I'll reserve judgement until after Neil's talk tonight.
PHP, with supporting libraries, is "good enough" to make sites from the
most simple to insanely complex. But as a language on it's on, it's a
piece of crap.
Nigel
Dean
perhaps you should go and join the ASP mailing list then instead :)
Jasper Bryant-Greene
like Nigel says, it's "good enough". It is still a crappy, poorly
architected language, but it works well enough for some applications.
Jasper
Alexei
Alexei
Jasper Bryant-Greene
The following of actual OO theory rather than the half-baked approach
PHP has taken.
The ability to actually get a release together without endless flamewars
on the developers' list.
Developers that at least partly care about the security of their
language/platform.
Sid Bachtiar
think PHP language is crappy at all. Infact, I think it's very good
for building websites. And I've used asp, asp.net, jsp, and perl to
compare it with.
I like it's flexibility. Sure, flexible language tends to make it easy
to make mistake, but that's what libraries and frameworks are for. We
shouldn't code from scratch everytime but rather use existing and
tested libraries and frameworks out there. Coding from scratch (or
reinventing the wheel) is error prone with any language out there.
It's good that PHP has many libraries and frameworks to suit each
individual's style and preference.
Sure the language needs to keep evolving and I think PHP 5 and 6 is a
proof that PHP is evolving for the better.
Alexei
i dont feel any lack of OOP. all only security problems i encountered
so far were caused by software developers, not language developers.
and its flexibility is great.
Alexei
Dave Lane
Having said that, given the huge amount of legacy code out there (we
have a 100k line app we recently ported to PHP5), it's impressive that
they've managed to give us the new OO capabilities and other features,
while simultaneously allowing us to port our code in our time rather
than theirs.
The alternative approach would've been to release a non-compatible
language called PHP that fixes all of PHP's architectural shortcomings,
and which would have been an unmitigated disaster. There's gotta be
some compromise. Some organisations have many person years invested in
development that they can't just upgrade at the drop of a hat - part of
what makes PHP compelling is its *user base* and developer communities.
Python is a better language, no question, but it's developer
community, while perhaps more competent on average (no offense to anyone
here - I'm not a Python developer myself) is tiny by comparison.
Anyway, to answer the original question - we (Egressive) provide
commercial PHP5 hosting (and have done for over a year - currently on
PHP 5.1.6 on Ubuntu 6.06) here in Chch, and we specialise in hosting
Drupal implementations (we also build a lot of them).
Cheers,
Dave
--
Dave Lane == Egressive Ltd == da...@egressive.com == +64 21 229 8147
+64 3 963 3733 = Linux: it just tastes better = no software patents
http://egressive.com ==== we only use open standards: http://w3.org
Effusion Group Founding Member =========== http://effusiongroup.com
Nigel McNie
> Well... if PHP is so bad...
> perhaps you should go and join the ASP mailing list then instead :)
ASP is worse. I might join the ruby or python lists though... not that
they don't have their weaknesses, but they're a lot better (imho) than PHP.
Nigel
Nigel McNie
>
> What do you like in other languages over php?
Consistent function naming/parameter order. Cleaner syntax. More
standard libraries (this is a biggie). Coded more securely. Devs that
aren't idiots. No commercial company driving it. Free license. No error
messages on line -1 billion. No problem with passing func_get_args() as
a parameter to another function call. Sane references. No magic_quotes
nightmare. No safe_mode hax.
I could go on, but I'll stop there.
Don't get me wrong, I use php every day. I'm just tired of working
around stupid design decisions made to support the bad old php3 days.
Nigel
Nigel McNie
>
> It's good that PHP has many libraries and frameworks to suit each
> individual's style and preference.
>
Hardly. Plenty of languages have more, better architectured libraries
which mean that you don't actually have to choose between N different
DBALs or similar for each project.
How would you parse a CSV file in PHP?
How would you access a Database?
How would you do templating?
CSV: As far as I've found, there's only one PEAR thing for it, and
that's old, not supported and buggy anyway. There's a function, fgetcsv,
but I can't remember what that does.
Database: PDO? PEAR? ADODB? Depends on what version you're using,
personal preference... decisions decisions that you shouldn't need to make.
Templating: Smarty pretty much wins. But plenty of people like using PHP
raw.
I argue, that on the contrary, it is _not_ good to have many libraries
to do these things. There's nothing that can be done about it now, but
if PHP would have had a standard library put in place that was
consistent, battle-hardened and flexible, people would not be writing
all these things themselves and wasting countless hours doing so.
Nigel
Nigel McNie
>
> yes, i agree. i dont think php is bad.
> i dont feel any lack of OOP. all only security problems i encountered
> so far were caused by software developers, not language developers.
> and its flexibility is great.
>
Have you not heard of the Month of PHP bugs, going on right now? Or of
the falling out of Steffen Esser with the PHP dev crew?
Nigel
Steve Boyd
to use.
If MacGyver was into programming im sure he'd use PHP ;)
> signature.asc
> 1KDownload
Cameron Junge
Stig Manning | Duco
He definitely would. PHP is especially good for mashing together many small
unlikely pieces of script to perform amazing feats; sure it might make
spaghetti code but hell, it works... right?
Berend de Boer
Hash: SHA1
>>>>> "Nigel" == Nigel McNie <oracle....@gmail.com> writes:
Nigel> I argue, that on the contrary, it is _not_ good to have
Nigel> many libraries to do these things.
Let's say I want to do something with AJAX. At last count there are
over 200 libraries. Is that bad?
I think many libraries are the inevitable result of progress. People
come up with better ideas, licenses don't suit all project, innovative
ideas mean a completely different approach is taken, etc. etc.
You start clean, and within 15 years, you're back to where you were.
- --
All the best,
Berend de Boer
PS: This email has been digitally signed if you wonder what the strange
characters are that your outdated email client displays.
PGP public key: http://www.pobox.com/~berend/berend-public-key.txt
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>
iD8DBQFF7iEaIyuuaiRyjTYRAnbIAKDaJuVY20sZcotMqH9ELx2BcvpLfgCggKb0
OX8mrm3eorwbwc11GNRU11w=
=AynM
-----END PGP SIGNATURE-----
Nigel McNie
>>>>>> "Nigel" == Nigel McNie <oracle....@gmail.com> writes:
>
> Nigel> I argue, that on the contrary, it is _not_ good to have
> Nigel> many libraries to do these things.
>
> Let's say I want to do something with AJAX. At last count there are
> over 200 libraries. Is that bad?
Yes. That's too many. Much too many. I can appreciate up to 5 different
ways of doing something, perhaps, maybe. But not 200.
> I think many libraries are the inevitable result of progress. People
> come up with better ideas, licenses don't suit all project, innovative
> ideas mean a completely different approach is taken, etc. etc.
Libraries are subject to the rules of evolution. 195 of those libraries
will fail. In the mean time, they clutter the landscape and make
choosing just one harder than necessary.
You think about the libraries that do ajax. What ones can you name off
the top of your head, and think that I could also name? Those are the
ones you should use. And the ones that the authors of the other 195
should be using, instead of scratching a pointless itch. (I'm assuming
here that the AJAX libraries, as in my experience, also bundle other
functionality and that your choice of which one is influenced by that.
For cases like DB access, there should only be one choice).
I'm all for innovation. I'm all for a new library that does something
nobody has done before, or satisfies a new niche, or solves an old
problem in a new, smarter/simpler/easier way. I'm not for all this
creation of libraries for the sake of it, that don't add any value. I
wish people would spend more time on the important things.
You're right about people coming up with new ideas being a good thing.
People re-inventing something with no advantages is not good.
Nigel
Citizen Gold
> Sid Bachtiar wrote:
> >
> > It's good that PHP has many libraries and frameworks to suit each
> > individual's style and preference.
> >
>
It's sounding like you would prefer a language where everything was
set in stone, there was no possibility of extending with
functions/classes/objects at all.
Kinda defeats the purpose of OOP doesn't it...
--
p: +6421 248-4653
w: http://www.babel.com.au/gold/
e: citize...@gmail.com
i : 92568615
Nigel McNie
>
> On 07/03/07, Nigel McNie <oracle....@gmail.com> wrote:
>> Sid Bachtiar wrote:
>> >
>> > It's good that PHP has many libraries and frameworks to suit each
>> > individual's style and preference.
>> >
>>
>> Hardly. [snip]
>
> It's sounding like you would prefer a language where everything was
> set in stone, there was no possibility of extending with
> functions/classes/objects at all.
>
> Kinda defeats the purpose of OOP doesn't it...
>
No, not at all. Extension to do what you want is fine. That's different
from re-inventing the wheel and you know it.
Citizen Gold
So you're anti-choice then?
rob.mc...@nzpages.net
> At dinner last night after Webstock Mini I came into a revelation that
> many New Zealand webhosts do not support PHP5 (I think ICONZ was
> mentioned as an example). I say this because we've been using PHP5-
> only features for quite some time, and moved our architecture from
> PHP4 to 5 to get its great new features (OO, stability, general new
> wad of functions, etc) back in 2005. We run our own servers so we can
> choose to keep up with the current stable architecture of Apache2,
> MySQL 5, PHP 5, etc.
And thats a bad thing?, the more people that dont support PHP5 just makes
it all the better for the hosts that do. :-) Stay with php4 I say, means
I'll make more money offering PHP5 to the people who value it!
:-D Rob
<!-- Snip snip -->
Grant Paton-Simpson
--
___________________________________
Dr Grant Paton-Simpson
Director, Paton-Simpson & Associates Ltd
16 Summit Drive, Mt Albert, Auckland 1025
(09) 849-6696
(09) 849-6699
___________________________________
anru chen
think about it, what kind of applications are wrote by php. etc,
oscommence, joomlar, symphony. all of those application is not so
called "large scale" software and users of those applications just
want to show some thing on the internet.
be honest, can adding OO to php help those user sell more products
through the internet?.
i am doing mantainence work for an oscommence site, OO or not OO is
not a big deal, old style procedure programming working very well.
unless, php can break into Luxury corporation world that actually
"burn" money, othwise just wait, one day (who knows) , php4 will gone.
regards,
anru
Grant Paton-Simpson
What specific aspects of the PHP5 OO model do you have issues with? I
am not challenging you or anything - I am genuinely interested. I
program in python as well and I want to improve my OO thinking.
All the best, Grant
P.S. I might be a bit too forgiving of PHP - I used to work a lot with
VBA in Access.
--
Nevyn
I've got to agree with Nigel here.
Take the most common tasks i.e. database interfacing - and standardise
them. Why are there 20 odd ways of doing the same thing when none of
them offer any benefits?
The first rule when programming something is to see if it's already
been done and if you can use it. If it has and it does everything you
want it to do, don't d!ck with it. If it doesn't, follow the open
source route and see if you can contribute code to make it do what you
need it to do. If that option isn't avaliable, then you have a point
of differentiation and it might just be worth coding your own.
You're all programmers right? Is this not what you do when you start a
new project?!?
Nigel McNie
>
> Let's face it, PHP isn't Python... but not every language can be :)
> Having said that, given the huge amount of legacy code out there (we
> have a 100k line app we recently ported to PHP5), it's impressive that
> they've managed to give us the new OO capabilities and other features,
> while simultaneously allowing us to port our code in our time rather
> than theirs.
>
> The alternative approach would've been to release a non-compatible
> language called PHP that fixes all of PHP's architectural shortcomings,
> and which would have been an unmitigated disaster. There's gotta be
> some compromise. Some organisations have many person years invested in
> development that they can't just upgrade at the drop of a hat - part of
> what makes PHP compelling is its *user base* and developer communities.
> Python is a better language, no question, but it's developer community,
> while perhaps more competent on average (no offense to anyone here - I'm
> not a Python developer myself) is tiny by comparison.
>
Given the slowness at which people have even begun to think about moving
to PHP5, I'm convinced that it wouldn't even be that bad. Look at
apache1.3 - 2.0. Heaps of sites still use apache 1.3. There's no reason
why PHP5 could not be maintained while people wrote _new_ applications
in this hypothetical "PHP" language.
In other words, it wouldn't be a language you "upgrade" to really. You
might do in your own time, as part of a rewrite.
Nigel
Sid Bachtiar
kinda reminding me of Javascript a little bit. Although its community
is so darn tiny (the blog last time I checked was full of spam).
Python has funny weird syntax. What's with the indentation? the
implicit self? Uh, and the way it handles privates? An OO language
without proper privates? Anyway, this is way OT now.
On 3/7/07, Nigel McNie <oracle....@gmail.com> wrote:
Jevon Wright
"Man, Smarty is so freaking bloated, I'm going to write my own templating system, it's going to be faster, and easier."
[three months later]
"Well I've got a basic framework done and it works OK. I just need to add this new feature for now..."
Six months later, they've written everything that Smarty can do, only with four times the internal complexity and slowness, achieving nothing.
Mind you, I've fallen into this trap myself.
Jevon
Sid Bachtiar
choices are available out there. If it's not for the people who put
their time and effort into trying out something that [they think]
offer something different, we probably wouldn't have all the good
tools today. It's about trying, sharing, and learning.
Ok, many of these projects will fail, discontinued, etc ... so what?
That's why we need to use our brain to decide which
libraries/frameworks/etc to use for our projects.
What next? one linux distro only?
Nigel McNie
I think you're missing the point. Choice Is Good. Excessive Choice Is
NOT Good. People not googling before embarking on Yet Another MySQL DBAL
is Not Good.
The linux distro thing is possibly another example of what I'm talking
about. There's Debian, Ubuntu, Red Hat, Mandriva, Slackware, SUSE and
other such "big names", which all do things differently or are targetted
at different audiences, and as such have carved out good market share.
Then there's distros like kororaa which demoed the XGL stuff, that had
their time in the sun because they were doing something that nobody had
done before. And then there's the millions of other distros that are
created, born and fail because their only reason for being is because
the author had a patch rejected by upstream and decided a new distro was
the only way they could get what they want.
My viewpoint is not, and has never been, that one choice is the best in
all situations. An example like Perl's DBI is one where it plainly is.
But I've never said that one choice is the be all and end all.
Nigel
Nick Taylor
That's something I always say, though it didn't take me 3 months to
write my system... it took about 1/2 a day. It's tiny, it works and it's
about 1/1000th the size of smarty. It can be extended to handle any
level of complexity... but I've never needed to extend it.
RE: 20 different systems doing the same thing... I'd say that was
probably a good thing on the level of ecological diversity. There is a
virtue to being able to make a clean start as well.
I like the coding in Codeignite. I absolutely hate the coding in Drupal
- well the last version I looked at anyway. Sure you can save time
by re-using other people's code (and there are applications like
PHPadmin that I wouldn't dream of trying to write myself) but simplicity
starts from 0. Sometimes its good just to start over with a clean slate.
> ------------------------------------------------------------------------
>
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.413 / Virus Database: 268.18.7/713 - Release Date: 07/03/2007
Nathan Cox
hosts that let you run php4 and php5 (besides maybe Rimu Host who let
you do whatever the hell you like)? My current host has "no plans to
support php5" and I'd really like to upgrade, but at the same time most
of the CMSes and whatnot out there like php4, and I'm not keen to try
replace them just yet.
It seems like some kind of hosting solution that lets me run different
sites or subdomains in different php versions would be handy.
Nathan
Steve Boyd
Nick how does you simple templating system work?
on a project last year just for fun i made really basic one based on
something i skimmed over in an old php3 book. essentially:
- do your logic
- import an .htm file which had things like <#--myvar--#> sprinkled
around it
- str_replaced $myvar into <#--myvar--#>
it worked great for what i was doing
Grant Paton-Simpson
Nick Taylor
> Nick how does you simple templating system work?
>
>
> on a project last year just for fun i made really basic one based on
> something i skimmed over in an old php3 book. essentially:
> - do your logic
> - import an .htm file which had things like <#--myvar--#> sprinkled
> around it
> - str_replaced $myvar into <#--myvar--#>
>
> it worked great for what i was doing
Yea - that's basically it:
foreach($tags as $k=>$v) {
$this->template=str_replace("[%".$k."%]",$v,$this->template);
}
And build from there. It can do loops, ifs, css-toggling (stripy
backgrounds in tables) and three or four other things like date_formating
(eg: [%startDate|date|d F Y%] - ie: uses php's formating codes)
smart nl2br (ie: doesn't put <br>s into tables or lists etc)
There's also a checkbox ticking function that I've never used - probably
because I haven't got the function naming simple enough yet. I've found
it's a fairly major step to get down to 1 syllable function names for
things that are used all the time.
so
loopHandler() becomes loop();
mergeTags() becomes merge();
etc
Another simple but major shift was using [% %] instead of <% %> - which
causes hassles when you're generating XML.
I use a separate function call for loops - The one in codeignite builds
the looped data into one big multi-level array with the one-off tags -
so the merge happens in a single hit.
I can sortof feel a temptation for going down this route... but it's
really useful having a templating system that can dovetail directly with
your DB class/models etc....
so you can have something (php-pseudocode) like
$clientData=$clients->getData();
$template->loop($clientData);
I'm still exploring table->class mapping. I can feel vague patterns
emerging - mostly to do with automatic form generation which will allow
quite dramatic condensation of code but they haven't quite crystalised
yet... and I'm far too busy writing emails in any case :)
n
Cameron Junge
from... This is very similar to what was being done in the very early
stages of PHP.
Kinda weird how history is repeating...
Cameron
-----Original Message-----
From: nzp...@googlegroups.com [mailto:nzp...@googlegroups.com] On
Behalf Of Steve Boyd
Sent: Thursday, 8 March 2007 11:24 a.m.
To: NZ PHP Users Group
Subject: [phpug] Re: PHP5 in New Zealand
Sid Bachtiar
Cameron Junge
to add dynamic info to webpages. Basically it was a simple templating
engine, and now it's a scripting language... Which is being used to
build templating systems!
Nick Taylor
I don't know if it's weird exactly - PHP was (and to a lot of people
still is) a templating language. All I'm doing is providing a more
user-friendly interface.
Or from a slightly different perspective - what I'm doing is writing a
language within a language - providing a much, much lower learning curve
than PHP itself has.
Sooo... theoretically, someone who knows a bit of HTML and has used
mailmerge can look at a template and immediately know what it does.
eg: [%client_name%] is a little less daunting than
<?PHP echo $this->client->["name"] ?>
Which will of course give an error which will be utterly terrifying to
anyone who isn't a programmer.
But most of the time I just do this for me... it just seems more elegant
and it reduces the number of keystrokes I need to make. It makes for
prettier and more predictable code.
Steve Boyd
half baked version of smarty =)
then again you can't call yourself a proper dev unless you have at
least 3 misguided projects that, at the time, were going to change the
world as we know it!
my 3:
- my half baked cms v1 (i tried doing WYSIWYG in flash 6 !!!!)
- my half baked cms v2 (php)
- thing im working on now
Nevyn
members of this user group? To share and collaborate? Ego can only
take you so far.
Nevyn
the most part, not trying to offer anything new. All they're doing is
cluttering up the landscape with crap which no one needs.
You'll find that most of the popular distros are those that can find
some point to differentiate themselves on.
I.e.
Debian - solid, stable but not cutting edge.
Ubuntu - Easy to use desktop
Novell Suse - Enterprise Level - taking on Microsoft users head on
Redhat Linux - Enterprise Level - old school favorite for enterprise solutions
What you need to understand is the word "different" (here's a hint,
it's the opposite of "same")
Sid Bachtiar
> What you need to understand is the word "different" (here's a hint,
> it's the opposite of "same")
Yeah, and I bet none of that 200 ajax php libraries mentioned above
are exactly the same. Otherwise it's just a file copy.
Aaron Dustin
Programmers and Designers (me included) moan louder than NZ Cricket
fans on talkback. Sure there may be reason for it, but I like how
doctors get on with, you know, like, saving peoples lives 16 hours a
day. And thats with lots of instruments to choose from ;)
Sorry, couldn't resist, as you were....
Nick Taylor
I'll bet you 874365837465 dollars if that's ok? :)
I've been deving this for 7 years now, and every time I rewrite it it
gets smaller. If I wind up doing a load of accountancy type stuff then I
might do a number formatting thing for it - but it will still be under
150 lines long if you take the comments out.
Know what you mean though - I went through a phase of making drag and
drop CMSs - this is the last surviving site
http://www.coyotevideo.co.uk - circa 2000
all done by dragging and dropping absolutely positioned divs on the fly.
I just know that sooner or later someone's going to throw a load of
money at this technique and try to do some sort of myspace type thing
with it. Not me though - although I did try that as well in about 2004.
Complicateder and complicatidider and complicatiderederder
Citizen Gold
>
> Yup, but my point is that PHP originally was about processing comments
> to add dynamic info to webpages. Basically it was a simple templating
> engine, and now it's a scripting language... Which is being used to
> build templating systems!
<mypetpeeve>
...templating systems with scripting languages built into them...
</mypetpeeve>
Just one of the reasons I think Smarty sucks the monkeys nuts.
--
p: +6421 248-4653
w: http://www.babel.com.au/gold/
e: citize...@gmail.com
i : 92568615
Jevon Wright
Jevon
Citizen Gold
>
> Because in Smarty you can optionally use PHP code?
No, because the way I've always looked upon templating systems is that
they're meant to _separate_ the business logic from the presentation
layer. Smarty has put the business logic back into the template by
adding their own scripting language (written in another scripting
language (how bloody retarded is that?)) for use in the templates.
--
p: +6421 248-4653
w: http://www.babel.com.au/gold/
e: citize...@gmail.com
i : 92568615
Suck my addiction...
Warren
>
> On 09/03/07, Jevon Wright <je...@jevon.org> wrote:
>>
>> Because in Smarty you can optionally use PHP code?
>
> No, because the way I've always looked upon templating systems is that
> they're meant to _separate_ the business logic from the presentation
> layer. Smarty has put the business logic back into the template by
> adding their own scripting language (written in another scripting
> language (how bloody retarded is that?)) for use in the templates.
I have utilised smarty on a few projects and it has helped in the long run
with keeping the business logic separate from the business logic - changes
have been able to be made incredibly quickly.
My worry has always been about speed - as CG mentioned above - a scripting
language, written in a scripting language ...
I always thought that perhaps a better way would be an actual PHP
module/plugin - written in something much faster (which essentially the way
PHP was written means C++, unless you start do wrapper stuff).
Although if it's a true template, it shouldn't have "logic" in it ...
Then, I started thinking about maybe an Apache module ... Something like
TSPPHP (Template system POST-PHP) ;-) ....but, whatever.
I still have issues with the way that "internet" applications are developing
anyhow. We have the technology and knowledge, maybe we need a shift away
from the browser paradigm and into something more akin to delivering the
"application" over the net and having that communicate back to the data
sources via XML. (Kinda like gtk/tcl scripting) ... But I digress somewhat.
My /2c.
Wazza.
Alexei
that code can be used for drawing graphs for example.
ie. business logic calculates & returns data and then template decides
how to draw it and uses its own code for it
Citizen Gold wrote:
>
> On 09/03/07, Jevon Wright <je...@jevon.org> wrote:
>>
>> Because in Smarty you can optionally use PHP code?
>
> No, because the way I've always looked upon templating systems is that
> they're meant to _separate_ the business logic from the presentation
> layer. Smarty has put the business logic back into the template by
> adding their own scripting language (written in another scripting
> language (how bloody retarded is that?)) for use in the templates.
>
--
Regards,
Alexei Tenitski
--
Net24 Limited
Web: www.net24.co.nz
Phone: 0800 5000 24
--
// domain names / email hosting / web hosting
// our reputation for reliability precedes us
Rob van der Linde
> I still have issues with the way that "internet" applications are developing
> anyhow. We have the technology and knowledge, maybe we need a shift away
> from the browser paradigm and into something more akin to delivering the
> "application" over the net and having that communicate back to the data
> sources via XML. (Kinda like gtk/tcl scripting) ... But I digress somewhat.
>
> My /2c.
> Wazza.
>
If the target browser of your web application only happened to be
Firefox/Mozilla (which is ok for internal web apps - we do this
ourselves), then XUL has been able to do this for years already,
checkout the "Mozilla Amazon Browser" as an example. Feels very
responsive, and the apps look native to the host OS.
Alexei
to use it as a table but later designer decided to generate table plus
graph to make data easier to understand (and page look nicer).
i didnt have to change a single line of business logic cause all what
was required for is a bit of 'presentation' logic.
Nick Taylor
So what - would you use a templating class to draw graphs?
I'd keep graph-drawing entirely seperate - there are a lot of third
party apps that do this. The trouble with having graph-drawing as a
templating thing is that graphs often need a hell of a lot of config
parameters - it'll wind up massively increasing the size and complexity
of your templating system - for something that isn't used that much.
I think I'd have graph-generation as a plugin to a report-generating
system rather than a templating system.
Warren
Thanks Rob,
I knew there was a proper answer to what I was looking for!
Cheers,
Warren.
Alexei
added without me having to do anything by editing template.
it was a simple graph though, using divs, no image generation stuff.
ability to have code in template gives greater separation between
business & presentation logic.
my templates generate page shortcuts, i just provide them with number of
pages and current page number. page shortcuts is not part of business
logic, it is presentation: maybe designer will feel like only next/prev
button, i dont care, i give him ability to do whatever is best.
that is why presentation layer should have code, different type of code.
does not matter if it is special smarty language or php or xslt as it is
in my case
Alexei
Citizen Gold
To put my interpretation of the term in context at any point where
there is a scripting language built into the template system and it is
up to the template designer to determine how and, more importantly,
_when_ certain block(s) of the template are to be parsed and displayed
the template system has moved out of the realm of being a template
system.
Joe Average web developer could not develop for the system without
first learning a new scripting language. A template system, IMHO,
should be able to be picked up by Joe Average web developer with a
very short list of rules to turn their html into a template.
Citizen Gold
>
> On 2007/03/9 10:24, "Citizen Gold" <citize...@gmail.com> wrote:
>
> >
> > On 09/03/07, Jevon Wright <je...@jevon.org> wrote:
> >>
> >> Because in Smarty you can optionally use PHP code?
> >
> > No, because the way I've always looked upon templating systems is that
> > they're meant to _separate_ the business logic from the presentation
> > layer. Smarty has put the business logic back into the template by
> > adding their own scripting language (written in another scripting
> > language (how bloody retarded is that?)) for use in the templates.
>
> I have utilised smarty on a few projects and it has helped in the long run
> with keeping the business logic separate from the business logic - changes
> have been able to be made incredibly quickly.
>
> My worry has always been about speed - as CG mentioned above - a scripting
> language, written in a scripting language ...
Glad I'm not the only one that thinks this way. :)
> I always thought that perhaps a better way would be an actual PHP
> module/plugin - written in something much faster (which essentially the way
> PHP was written means C++, unless you start do wrapper stuff).
Agreed. Smarty implemented as a PECL module would make a hell of a
lot more sense.
> Although if it's a true template, it shouldn't have "logic" in it ...
Thank you. :)
> Then, I started thinking about maybe an Apache module ... Something like
> TSPPHP (Template system POST-PHP) ;-) ....but, whatever.
How would you see that working?
> I still have issues with the way that "internet" applications are developing
> anyhow. We have the technology and knowledge, maybe we need a shift away
> from the browser paradigm and into something more akin to delivering the
> "application" over the net and having that communicate back to the data
> sources via XML. (Kinda like gtk/tcl scripting) ... But I digress somewhat.
XUL. There was a post prior to this reply suggesting it for firefox
but there's no reason you couldn't make a Gecko based app (like
Firefox/Thunderbird/Seamonkey/Camino/Nvu/Sunbird/etc) that is a
standalone application and doesn't need FF as a prerequisite. Also,
with very little work, you get a cross-platform application.
Warren
>
>> I always thought that perhaps a better way would be an actual PHP
>> module/plugin - written in something much faster (which essentially the way
>> PHP was written means C++, unless you start do wrapper stuff).
>
> Agreed. Smarty implemented as a PECL module would make a hell of a
> lot more sense.
>
I did have a little play with trying to make a PECL module once ... The
problem more was compiling it for all the differing systems out there...
>> Then, I started thinking about maybe an Apache module ... Something like
>> TSPPHP (Template system POST-PHP) ;-) ....but, whatever.
>
> How would you see that working?
I didn't really - it was almost tongue in cheek... But, more along the
lines of the PHP/whatever delivering the "data", and the system knowing
which template to pick up, and merging the two before being delivered to the
web browser. Although I'll be the first to admit I really haven't thought
this fully through ...
>
>> I still have issues with the way that "internet" applications are developing
>> anyhow. We have the technology and knowledge, maybe we need a shift away
>> from the browser paradigm and into something more akin to delivering the
>> "application" over the net and having that communicate back to the data
>> sources via XML. (Kinda like gtk/tcl scripting) ... But I digress somewhat.
>
> XUL. There was a post prior to this reply suggesting it for firefox
> but there's no reason you couldn't make a Gecko based app (like
> Firefox/Thunderbird/Seamonkey/Camino/Nvu/Sunbird/etc) that is a
> standalone application and doesn't need FF as a prerequisite. Also,
> with very little work, you get a cross-platform application.
All the times I've been thinking of using something like this, it's more for
our clients / administrators rather than joe public looking at websites, so
I don't see any major problems telling people they must use Firefox for the
administration.
Plus, given the limited amount I've looked into it, then I think it should
be possible to simply open in a new window and make the whole thing look
like an application. There are a couple of things I'd like - one is a
recent RTE area (which we all know is already in FF) and the ability to
auto-fill or use a drop down box that can also have extra items typed into
it. This, treated properly would minimise the requirements for creating
form based systems.
Ugh - I rant too much... But there's much that I must look into now :-)
Wazza.
Mark @ intelliServe
Just looking for some recommendations for an application that I can
paste/open complete files into that will format the php. If possible any
html as well. Just looking to tidy up completed work.
Thanks.
Regards
Mark
Timaru
James McGlinn
> Just looking for some recommendations for an application that I can
> paste/open complete files into that will format the php. If
> possible any
> html as well. Just looking to tidy up completed work.
You may want to repost this with a new thread. Because you replied
to the 'PHP5 in New Zealand' messages people with that thread closed
(in threaded mail readers) won't see it.
http://en.wikipedia.org/wiki/Thread_hijacking
Kind regards,
James McGlinn
BCom, BSc, Zend Certified Engineer (PHP4, PHP5)
Director
Nerds Inc.
The Loft, 201 Hobson St, Auckland
(09) 377 5663 / (021) 633 234
http://www.nerdsinc.co.nz http://blog.phpdeveloper.co.nz
Anthony Gallon
----- Original Message -----
From: "Nevyn" <nev...@gmail.com>
To: <nzp...@googlegroups.com>
Sent: Thursday, March 08, 2007 1:20 AM
Subject: [phpug] Re: PHP5 in New Zealand
>
Citizen Gold
http://googlegroups.com and take it from there.
sigurdm...@gmail.com
Nucleus. They now support PHP5 in new accounts created with them, and
their infrastructure is physically located in Auckland, rather than in
the states. I've briefly confirmed that you can set up a SilverStripe
site on it ...
Alexei
> Nucleus. They now support PHP5 in new accounts created with them, and
> their infrastructure is physically located in Auckland, rather than in
> the states.
We support PHP5 too now. Servers are in our own datacenter in
Christchurch (http://net24.co.nz/network.htm).
Ryan McCarvill
Steve Boyd
So does iserve - for some reason they haven't updated their promotional
stuff to reflect this.
Also support Mysql 5
Though you don't get much HD space unless you want to pay extra (20 or
25 meg for your $20 a month). Also they restrict things like
file_get_contents('http://www.whatever.com');
Good host, had a site with them forever
Sarah King
right now.
I love some of the folder names, eg:
/cda/tvnz/ugly_content_feature_box_skin/
Sarah
sigurdm...@gmail.com
their website :P
I am going to be interviewed tonight on CloseUp (TVOne 7pm) to explain
how an open source project can make money! :)
Siggy
Sarah King
I'll be out but will set my HDD to record.
Good luck!
You know that coffee and cars quote that was doing the rounds last week?
http://www.andrewking.co.nz/pm/don-t-buy-coffee-and-cars-if-you-want-a-house
/
You need to come up with your own version for the corporate world...
Feel free to mention that the site uses SilverStripe ;)
Sarah
-----Original Message-----
From: nzp...@googlegroups.com [mailto:nzp...@googlegroups.com] On Behalf
sigurdm...@gmail.com
you're definitely on my list :)
anru chen
regards,
anru
Berend de Boer
Hash: SHA1
>>>>> "sigurd@silverstripe" == sigurd@silverstripe com <sigurdm...@gmail.com> writes:
sigurd@silverstripe> I am going to be interviewed tonight on
sigurd@silverstripe> CloseUp (TVOne 7pm) to explain how an open
sigurd@silverstripe> source project can make money! :)
Can or has?
:-)
- --
All the best,
Berend de Boer
PS: This email has been digitally signed if you wonder what the strange
characters are that your outdated email client displays.
PGP public key: http://www.pobox.com/~berend/berend-public-key.txt
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>
iD8DBQFGNr/MIyuuaiRyjTYRAt9TAJ9uhx7KeUq3iNLAyHYhD4VL4w1MTACfUYWp
B9MEom9/O2MtSpG38Nd64+I=
=8L+L
-----END PGP SIGNATURE-----
Michael Brandon
Now who is in for creating a rehashed item via the closeup website??? Will
be fascinated in seeing the interest that it generates.
Regards
Michael
Kieran P
From the TVNZ website:
http://tvnz.co.nz/view/page/497100/1090817
I missed the show tonight, so I cant wait to check it out :D
Kieran.P
http://qlwiki.linuxsolutions.co.nz/
sigurdm...@gmail.com
source. Its great to have a prime time newsman (Mark Sainsbury)
expressing the virtues of open source to a large audience.
Definately helps us to pitch the idea to people!
I found some relevant TVNZ webpages
Overview: http://tvnz.co.nz/view/page/497100/1090817
Video: http://tvnz.co.nz/view/video_popup_flash_skin/1090868
And I'll be sure to tell a few people about mashing up the piece ! :)
sigurdm...@gmail.com
On May 1, 4:19 pm, Berend de Boer <ber...@pobox.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> >>>>> "sigurd@silverstripe" == sigurd@silverstripe com <sigurdmagnus...@gmail.com> writes:
>
> sigurd@silverstripe> I am going to be interviewed tonight on
> sigurd@silverstripe> CloseUp (TVOne 7pm) to explain how an open
> sigurd@silverstripe> source project can make money! :)
>
> Can or has?
>
After watching the start of CloseUp (on smacking), before I was on, I
was reminded how typically no one answers the question actually asked.
So I'll answer with "is" :P
Seriously, though, things are good. Being in front of more eyes means
more potential paying customers, in the very same way that allows
everyone from MySQL to RedHat to operate. I've already had a few
emails this evening with new people interesting in talking business.
Secondly, to most people, the open source practice creates a curiosity
about us and that's really good too. I doubt we'd be on CloseUp had we
written the appliacation in .Net and sold it for USD100,000 ....
Harvey Kane
I thought you said your piece well without making us all look too much
like cyber-hippies.
The other guy at the end of the article was missing the point though -
people will spend thousands of hours editing a film or app because they
want to - not because there is some guaranteed financial reward at the
end. The really good stuff on the web comes from creative people trying
new things and seeing what happens, and while a lot of ideas come to
nothing, the odd one goes gold.
Who would have thought you could build a web based encyclopedia and let
people edit the content without a password? - what a fruity idea :)
Harvey.
--
Harvey Kane
Web Development & SEO Specialist
mb 021 211 1951
ph +649 838 1331
www.ragepank.com
Dennis Smith
I was all confident that I had it on Mysky but . . .
My Mysky has recorded the last 326 episodes of Close-up perfectly then
dropped a whole bunch of settings including your Close Up one tonight
It does that occasionally - reboot and reprogramme is the only way I've
worked out how to fix it
I guess they must use a MS operating system inside their set-top
Regards . . .
Dennis Smith <http://www.wdanz.co.nz>
Dennis Smith, CEO
Web Developers Association of New Zealand Ltd
_____
Ph +64 9 486-6410 ~ Fax +64 9 489-3077 ~ Mobile: 021 WEBSITES
P O Box 101-470 ~ North Shore 0745 ~ www.wdanz.co.nz
<http://www.wdanz.co.nz>
_____
-----Original Message-----
From: nzp...@googlegroups.com [mailto:nzp...@googlegroups.com]On
Behalf Of sig...@silverstripe.com
Sent: Tuesday, 1 May 2007 2:07 p.m.
To: NZ PHP Users Group
Subject: [phpug] Re: [OT]TVNZ
sigurdm...@gmail.com
few hours ago.
Overview: http://tvnz.co.nz/view/page/497100/1090817
Video: http://tvnz.co.nz/view/video_popup_flash_skin/1090868
Neil Bertram
server:
http://youtube.com/watch?v=y068GJ1OEC8
http://www.webbedfeet.net.nz/stuff/TvOne-CloseUp-OpenSource.avi
Neil
sigurdm...@gmail.com
anru chen
i just saw that video today, but i have no speaker here in my working
place, so no voice for me :(, i will check it later tonight when i am
home.
but i am pretty much dislike one video segment that a guy got a free
car. (since i have no speaker here, i hope my understanding is right).
that a really bad analog.
regards,
anru
Chris Hope
source, by saying that open source was like a car manufacturer giving
you away the car and all the manuals, blueprints etc. The analogy would
have been OK if they didn't talk about giving the car away - that would
be like saying with open source you get a free server along with your
free operating system.
Cheers,
Chris
sigurdm...@gmail.com
you can convey in a few sentances, so there HAS to be some detail
omitted. An analogy, by definition, is never a substitute for truely
understanding the actual topic.
People often say free software is free as in speech, not beer. I'm
assuming here you think the car is too much like beer, and I'd suggest
otherwise. Beer can only be consumed, i.e. its hard for people to
think of modifying the beer that is given to you. A car, on the other
hand, is frequently directly modified/enhanced after being
purchased... just think of boyracers... so the general public can
identify, through car modifications, that free software is then also
editable.
Anyway, enough of that... let's be happy the word got out at all :)
Aaron Cooper
> People often say free software is free as in speech, not beer.
Or free as in 'pirated' in some dodgey basement, burned onto 40 cent Blank
Princo CDs............