Backend on the internet?

Danny J. Lesandrini wrote:

Well, you're correct that by converting to an ADP he could link to his
data, but that would require purchasing SQL Server (or possibly
installing MSDE) and opening up port 1433. I've done this, in testing
only, and the performance was nearly tolerable where I wanted to
retain all the front end functionality. So, if that is his goal, to keep
all his front end, then you're right, he could go ADP and SQL Server.

However, if all he wanted was to get to the data via the internet, then
it's very easy to do ASP. I thought you were disparaging ASP as a way
to access data via the web. (I've heard it said that MDBs don't perform
well with web apps, but I don't believe it.)

Danny

I have used both ADPs and ASP to connect to rented space(100 meg
database) on an Internet enabled MS-SQL server.
The ADP's performance is indistinguishable from performance when it is
attached to a local, on the same machine, MS-SQL server.

You and I both seem to be fans of classic ASP. It's fabulous in my opinion.

<unprofessional rant>
I have used both Interland.Com and DiscountAsp.Net for this. I believe
Interland is bigger and flashier. It also cost me three times what
DiscountAsp cost. Interland was worth the extra money, I guess, because
it had three times the problems, three times the number of loss of
service occasions, and three times the trouble in getting anything fixed
up. I left Interland when it wrote other's SQL USER IDs and Passwords to
my Web Server space and refused to recognize that this was a security
problem, and when they quoted me a high price for retrieving a backup
copy of my db. I'm a fan of DiscountAsp.Net and have three accounts with
them; [No, I don't have shares in the company and have not received any
favours from the CEO's daughter.] (Did I mention three times the
problems with E-mail?).
</rant>

TTBOMK port 1433 may be more dangerous only because it is the default
that MS-SQL Server uses and thus, generally the first that an SQL hacker
tries. I have connected to a wholly owned and dedicated Internet enabled
MS-SQL server on a port other than 1433 (was in the 5700's I think). I
noticed no difference in performance or security from the rented 1433
ones, except that DiscountAsp's run without any problems or tinkering.
The big advantage, maybe, to the independent MS-SQL server was that we
could administer the whole server, but it was not faster than the rented
dbs. Because we could administer it, we did administer it; I'm not sure
this was a good idea; it certainly had much more problems that the
rented one. In getting the data for this server there was a
(confession!) time where I got impatient and WAS FORCED TO!!!!!! hack
the db on the previous supplier's server to get the data to meet a
deadline. The 1433 port was handy, I guess, but if it hadn't been 1433 I
would just have kept trying programmatically until I got one that
worked. I suppose security on the server might have noticed repeated
attempted incursions from my IP address and shut me off but nothing like
that happened.

In any case, all the SQL servers I have seen (except those of
DiscountAsp where conditions are more strict), (probably fewer than
thirty) have inadequate security enabled at the SA level and I have been
able to do anything I want with any db I want. I really haven't tried at
DiscountASP, happy to believe that it can't be done.

But, as you may know, I am not a fan of ADPs. One reason is security.
Another is the number of connections they make to the server. Perhaps
this was peculiar to the occasion but a year or so ago I introduced and
ADP to an organization. We had about 40 users with laptops using their
own copy of the ADP. The System administrator who was present was quite
curt and I couldn't understand why. Then he showed me his laptop screen;
my 40 ADPs accounted for almost 500 [five hundred] connections to the
Server. Perhaps, this was dependent on the fact that they were new to
the application and feeling their way. Perhaps it was due to my
programming style. Perhaps it was something else. But he didn't like it
and I couldn't explain it.

I think renting a 100 meg db on a modern (2000 or 2005) SQL Internet
enabled server at $40 USD per quarter is a hell of a deal. As I said, I
have three of them and I am very happy with them. I use ASP or Horrors!
HTA which are closely related as front ends for them and the people who
use these are très happy as well.

I'm rambling on ... time for lunch ... shut up Lyle!
--
Lyle Fairfield

Dec 16 '05 #9

Danny J. Lesandrini

Lyle, I was speaking from very limited experience with ADPs. I agree with
you (though I hope Aaron doesn't notice this post) that I prefer MDBs. My
one example of using an ADP to connect to a remote SQL Server via Internet
proved workable, but not indistinguishable, though that may be due to the
size of my server. (In this case mine is probably not bigger than yours.)

As for Classic ASP, I am a fan. I created an app for the Colorado Division
of Wildlife for maintaining a list of properties they might like to purchase,
should they ever get the money. Each property entry logged things like the
kind of animals and fish to be found on them, road access and water rights.
They insisted on going Dot Net, so I obliged. Problem is, as you can imagine,
data entries are made to this database only about 4 times a month. Each time,
the Dot Net app had to take an incredible hit to load the CLR and all the
appropriate linked libraries. It was interminably slow ... for the first user ...
who happened to also be the last user before the entire works was unloaded.

Should have written it in Classic ASP.
--

Danny J. Lesandrini
dl*********@hotmail.com
http://amazecreations.com/datafast
"Lyle Fairfield" <ly***********@aim.com> wrote ...

The ADP's performance is indistinguishable from performance when it is attached to a local, on the same machine,
MS-SQL server.

You and I both seem to be fans of classic ASP. It's fabulous in my opinion.
--
Lyle Fairfield

Dec 16 '05 #10

Steve

I checked out http://www.discountasp.com and now wish I had known of
them before. My current host is okay right now, but when I first got
started with them it took three weeks to get everything set up
correctly. Hostony is based in Poland or some such and the support
personnel do not all have a good grasp of the English language. The
service is pretty unprofessional, as is the system it is served on. The
only things I needed them to do is enable IIS for ASP.NET support
(which was supposed to be already set up) and give the ASP.NET machine
account read/write access to one particular folder. Three weeks later,
the ASP.NET machine account has read/write to my ENTIRE SHARE (I tried.
I really did. After 8 repeated attempts to communicate "only this
folder", I gave up and let the support supervisor warn me about the
dangers of giving read/write to the entire share). Yeah, it'll get
trashed eventually, but at least it works.

However, Discount ASP seems to have good prices and good packages, and
if you're recommending them, I'll put my next web project on Discount
ASP servers.

Dec 16 '05 #11

Steve wrote:

However, Discount ASP seems to have good prices and good packages, and
if you're recommending them, I'll put my next web project on Discount
ASP servers.

Oh the pressure of it all! Will I have dyspepsia on Christmas Eve?

--
Lyle Fairfield

Dec 16 '05 #12

"Danny J. Lesandrini" <dl*********@hotmail.com> wrote in
news:N8********************@giganews.com:

Well, you're correct that by converting to an ADP he could link to
his data, but that would require purchasing SQL Server (or
possibly installing MSDE) and opening up port 1433. I've done
this, in testing only, and the performance was nearly tolerable
where I wanted to retain all the front end functionality. So, if
that is his goal, to keep all his front end, then you're right, he
could go ADP and SQL Server.

Is it really necessary to use an ADP? Can't you use ADO or ODBC over
the Internet from an MDB, too?

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 16 '05 #13

"GAVO-UK" <an****@peopleex.com> wrote in
news:11**********************@g47g2000cwa.googlegr oups.com:

Now that we are on the ADP, SQL subject and you mentioned port
1433. I understand that port 1433 creates a hole in the security.
Do you know an alternative port of security method to prevent
unauthorised access to the DB.??

Run the app across a VPN connection. D'oh.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 16 '05 #14

Shiva

So, the answer depends on whether he wants to access some data or to
use his entire front end.
--

Danny J. Lesandrini
dl*********@hotmail.com
http://amazecreations.com/datafast
"GAVO-UK" <an****@peopleex.com> wrote ...
As I understand from Shiva's question, the db has been split into a
front end and back end and now he/she wants to access the back end on
the net using the front end. I believe an ADP is the right step.

GAVO.

Thanks for all the replies. What I really want? Well, viewing the data of
the .mdb database is the only thing I want. So that would make ASP the most
obvious choice?
On the other hand, I don't know a thing about ASP nor ADP and I don't want
to spend too much time on creating new frontends by using ASP or ADP. Which
of these choices would the easiest to implement? Bear in mind that I already
got a frontend and I hope(d) I could use this somehow..

Dec 17 '05 #15

ADPs connect to MS-SQL server. They're easier. But I think you want to
connect to JET. So, I think they are out.

ASP requires scripting skills.Should you learn it? It's a favourite of
many but it's old, old, old.

Of course, new, new, new is ASP.net.

Viewing data is different from editing data. Until you confirm viewing
or editing it's difficult to give you a good answer as to what may be
best.

Access is a RAD platform. Using its wizards and user interface
beginners and dabblers can create quite sophisticated applications. But
Access does not develop internet applications, except in limited
fashion as Data Access Pages, and Access Data Projects (not everyone
would call these Intetnet Applications), and these two technologies are
not universally admired by experienced developers.

Spending too much time each day for thousands of days is what the
regular responders and experts have done. That is how they became
experts.

I think you need to define very carefully exactly what you want to do.
And then you need to spend a couple of Google days researching how you
might do it. After that you may have some other questions about the
specifics of your solution.

If there were a magic way of putting an Access application on the
Internet one might see many Access applications on the Internet. I have
seen none.

If there were a magic way of putting an Access application on the
Internet the owners of large sites wouldn't pay millions for site
development. I think they do.

Dec 17 '05 #16

"Lyle Fairfield" <ly***********@aim.com> wrote

[helpful stuff]

.... and that kind of reasoned, intelligent response is exactly why I still
read this NG despite all the spam.

--
Darryl Kerkeslager

Dec 17 '05 #17

rkc

Darryl Kerkeslager wrote:

"Lyle Fairfield" <ly***********@aim.com> wrote

[helpful stuff]

... and that kind of reasoned, intelligent response is exactly why I still
read this NG despite all the spam.

I was scoffed at the last time I brought this up, but there's no reason
to fuss around with CSS, HTML, and Javascript to get at .mdb data via
a web server. You can create your asp data objects on the server and
request information from them via the MS HttpRequest object or any other
means you see fit from within Access. You have to go unbound, but it's
way, way and way again easier and faster then creating a web interface.

Dec 17 '05 #18

Could you, please, give (or direct us to) an example.
(I promise to ride point against any scofffers!)

Dec 17 '05 #19

rkc

Lyle Fairfield wrote:

Could you, please, give (or direct us to) an example.
(I promise to ride point against any scofffers!)

I'll email you a simple proof of concept .mdb, but it's
nothing you couldn't come up with with a minimum of thought.

Dec 17 '05 #20

Waiting breathlessly!

Dec 17 '05 #21

So it seems that with this method, your class, one can grab the output
of any web document as a text/xml stream.
Stop me if I'm wrong:
If one designed ASP with this in mind one could download strings quite
efficiently. One could display them effectively as you do in your form.
So whatever the ASP could use, the MDB could use, including records
from another MDB on the web server.
Fascinating and not scoffable at all.
I didn't see anything about this but I'm guessing this is a one way
read and not an editing thing?
Or can we send the page back so to speak and post/get the form to save
the changes?

Dec 17 '05 #22

Rick Brandt

Lyle Fairfield wrote:

So it seems that with this method, your class, one can grab the output
of any web document as a text/xml stream.
Stop me if I'm wrong:
If one designed ASP with this in mind one could download strings quite
efficiently. One could display them effectively as you do in your
form. So whatever the ASP could use, the MDB could use, including
records from another MDB on the web server.
Fascinating and not scoffable at all.
I didn't see anything about this but I'm guessing this is a one way
read and not an editing thing?
Or can we send the page back so to speak and post/get the form to save
the changes?

I have been doing similar stuff for a few years now except that I post to a java
servlet running on our web server instead of using ASP. It works very well and
is very secure.

In my case I have never sent an actual SQL string to the servlet to be processed
but I see no reason why that wouldn't work. You would obviously need to filter
against action type SQL that could do something malicious. What I do is provide
the type of request and the user's credentials in the HTTPRequest and then
return text/xml in the response. Then it is just a matter of parsing the
response and putting it into the tables (or doing whatever else with it).

The same functionality could be used to get stock quotes, weather or map data,
etc., from just about any web service on the internet as long as you know what
to POST and the format of what will be returned.
--
I don't check the Email account attached
to this message. Send instead to...
RBrandt at Hunter dot com

Dec 17 '05 #23

rkc

Lyle Fairfield wrote:

So it seems that with this method, your class, one can grab the output
of any web document as a text/xml stream.
My class is just a simple wrapper, a facade in design pattern lingo,
around the Microsoft.XMLHTTP object. That was done so I could switch
out the XMLHTTP code for another method such as the api code that was
posted here a while back if need be.
Stop me if I'm wrong:
If one designed ASP with this in mind one could download strings quite
efficiently. One could display them effectively as you do in your form.
So whatever the ASP could use, the MDB could use, including records
from another MDB on the web server.
Yes. The customerID is sent to the server as a simple url encoded
name=value pair. The asp code on the server opens a recordset from an
..mdb file and sends it to the client in csv format. It could just as
easily send it as xml, which would probably make it easier to open as
an ado.recordset.
Fascinating and not scoffable at all.
I didn't see anything about this but I'm guessing this is a one way
read and not an editing thing?
Or can we send the page back so to speak and post/get the form to save
the changes?

You can, as you know, create a disconnected or fabricated recordset and
base an Access form on it. That would allow you to browse, and edit
multiple records as well as add new ones using forms. That does not, of
course, send the changes back to the server. You have to track the
changes and do updates via another post to the appropriate asp page on
the server. You can do that one record at a time via name=value pairs
or in a batch by posting back .csv or xml formated text as an
update=xmlText pair and having the asp script parse it.

Dec 17 '05 #24

rkc wrote:

Lyle Fairfield wrote:
So it seems that with this method, your class, one can grab the output
of any web document as a text/xml stream.

My class is just a simple wrapper, a facade in design pattern lingo,
around the Microsoft.XMLHTTP object. That was done so I could switch
out the XMLHTTP code for another method such as the api code that was
posted here a while back if need be.
Stop me if I'm wrong:
If one designed ASP with this in mind one could download strings quite
efficiently. One could display them effectively as you do in your form.
So whatever the ASP could use, the MDB could use, including records
from another MDB on the web server.

Yes. The customerID is sent to the server as a simple url encoded
name=value pair. The asp code on the server opens a recordset from an
.mdb file and sends it to the client in csv format. It could just as
easily send it as xml, which would probably make it easier to open as
an ado.recordset.
Fascinating and not scoffable at all.
I didn't see anything about this but I'm guessing this is a one way
read and not an editing thing?
Or can we send the page back so to speak and post/get the form to save
the changes?

You can, as you know, create a disconnected or fabricated recordset and
base an Access form on it. That would allow you to browse, and edit
multiple records as well as add new ones using forms. That does not, of
course, send the changes back to the server. You have to track the
changes and do updates via another post to the appropriate asp page on
the server. You can do that one record at a time via name=value pairs
or in a batch by posting back .csv or xml formated text as an
update=xmlText pair and having the asp script parse it.

The whole thing sounds fabulous to me!

--
Lyle Fairfield

Dec 17 '05 #25

"Darryl Kerkeslager" <ke*********@comcast.net> wrote in
news:kf********************@comcast.com:

"Lyle Fairfield" <ly***********@aim.com> wrote

[helpful stuff]

... and that kind of reasoned, intelligent response is exactly why
I still read this NG despite all the spam.

What spam?

I don't see any spam in this newsgroup.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 17 '05 #26

> "Darryl Kerkeslager" <ke*********@comcast.net> wrote

... and that kind of reasoned, intelligent response is exactly why
I still read this NG despite all the spam.

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote What spam?

I don't see any spam in this newsgroup.

What spam?? Surely you haven't blocked not only PCDatasheet, but every
other purveyor of their own software, their own agenda, and their own OT
interests ...

--
Darryl Kerkeslager

Dec 17 '05 #27

"Darryl Kerkeslager" <ke*********@comcast.net> wrote in
news:f8******************************@comcast.com:

"Darryl Kerkeslager" <ke*********@comcast.net> wrote
... and that kind of reasoned, intelligent response is exactly
why I still read this NG despite all the spam.

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote
What spam?

I don't see any spam in this newsgroup.

What spam?? Surely you haven't blocked not only PCDatasheet, but
every other purveyor of their own software, their own agenda, and
their own OT interests ...

I don't see any such things.

Are these crossposts? I don't see those, because I have my
newsreader set to kill any post with more than two newsgrouss.

But I don't see much spam, maybe one post per day. And if it weren't
for the responses of those pointing out that it's not allowed, I
would likely not even notice those.

And those are all drive-by posts -- the people making them will
never see the responses, and don't care. They don't read the
newsgroup and never will.

So, I simply don't think there's any significant spam problem in
this newsgroup. If you disagree, then perhaps you can provide Google
MessageIDs of all the numerous spam posts for, say, a single day of
your choice (in the last two weeks).

My guess is that even if you work very hard, you'll not find a
single day where the number of spam posts exceeds five.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 19 '05 #28

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote

Are these crossposts? I don't see those, because I have my
newsreader set to kill any post with more than two newsgrouss.
I use a low-feature news reader. I don't think I can do that.

And those are all drive-by posts -- the people making them will
never see the responses, and don't care. They don't read the
newsgroup and never will.
The spam in my email is "drive-by" also - but still spam.
So, I simply don't think there's any significant spam problem in
this newsgroup. If you disagree, then perhaps you can provide Google
MessageIDs of all the numerous spam posts for, say, a single day of
your choice (in the last two weeks).

How does Google figure into this? They don't own usenet.

--
Darryl Kerkeslager

Dec 19 '05 #29

Saintor

Frontpage might help you easily if managing Frontpage extensions is not an
obstacle.
"Shiva" <sh***@gonzo.com> wrote in message
news:dn**********@brutus.eur.nl...

Hi,

My access application has been split into a frontend and a backend. The
backend is on my harddisk. Now I'd like to put the backend on my homepage
(thus the internet) - this would allow me to access my data from any place
connected to the internet.
Before trying to set this up, I'd like to know if this is possible and if
it's a good idea anyway. If not, what alternatives do you have?

Thanks a lot.

Dec 19 '05 #30

"Darryl Kerkeslager" <ke*********@comcast.net> wrote in
news:3v********************@comcast.com:

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote
Are these crossposts? I don't see those, because I have my
newsreader set to kill any post with more than two newsgrouss.

I use a low-feature news reader. I don't think I can do that.

Well, then that's *your* fault -- get a decent newsreader.

Recently I ran across a new reason why combined email/news clients
are a bad idea: the rules for content encoding in newsgroup posts
and in email messages are COMPLETELY DIFFERENT, and controlled by
different RFCs. That means that one of two things happens:

1. the email/news client supports only one of them (likely mail)
with the result that it makes non-compliant posts to Usenet, OR

2. the email/news client's programmers have to support two
completely different posting formats, with all the extra work and
potential problems that entails.

I don't know what Outlook Express does. Indeed, I can't understand
why anyone with an IQ of greater than 60 would use Outlook Express
as, aside from the inherent insecurity of it, it's slow and has a
hideous UI.

If you like the combined route, Thunderbird is good.

If you want a real news reader, with everything that news readers
have had since the days of reading News in a Telnet window, try
xNews.

And those are all drive-by posts -- the people making them will
never see the responses, and don't care. They don't read the
newsgroup and never will.

The spam in my email is "drive-by" also - but still spam.

Are we talking about your email or about CDMA?

Additionally, the observation does not change my point: the spam
that comes into your inbox is not under your control, any more than
the spam that comes into the newsgroup. You're not receiving it in
your email inbox because of the content of your email messages, but
just because some spammer got hold of your address.

Similarly, what spam is coming into this newsgroup is not arriving
here as a result of a human being seeing "Aha! I see PC Datasheet is
advertising in his sig, so it's obviously OK for me to spam this
newsgroup with my advertisments! Nope, what is happening is that the
Unenet spammer sees CDMA in the list of newsgroups and sends the
message to the group.

So, as I said, what spam there is in CDMA is completely unrelated to
the behavior of individuals within this newsgroup. Steve's posts are
*not* going to open the floodgates of spam into this newsgroup.

So, I simply don't think there's any significant spam problem in
this newsgroup. If you disagree, then perhaps you can provide
Google MessageIDs of all the numerous spam posts for, say, a
single day of your choice (in the last two weeks).

How does Google figure into this? They don't own usenet.

Huh?

All I'm asking for is evidence that CDMA has a spam problem. Google
Groups should allow you to document that, if the problem actually
does exist.

Otherwise, you must admit that there is no such spam problem,
contrary to your allegation.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 19 '05 #31

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote

Otherwise, you must admit that there is no such spam problem,
contrary to your allegation.

Nuh-uh.

Read:

Spam refers to electronic junk mail or junk newsgroup postings. Some people
define spam even more generally as any unsolicited e-mail. In addition to
being a nuisance, spam also eats up a lot of network bandwidth. Because the
Internet is a public network, little can be done to prevent spam, just as it
is impossible to prevent junk mail. However, the use of software filters in
e-mail programs can be used to remove most spam sent through e-mail.

Or:

To indiscriminately send unsolicited, unwanted, irrelevant, or inappropriate
messages, especially commercial advertising in mass quantities. Noun:
electronic "junk mail".

Or:

(or Spamming) An inappropriate attempt to use a mailing list, or USENET or
other networked communications facility as if it was a broadcast medium
(which it is not) by sending the same message to a large number of people
who didn't ask for it. The term probably comes from a famous Monty Python
skit which featured the word spam repeated over and over. ...

Or:

As a noun, an irrelevant message (frequently commercial advertising)
cross-posted to many public fora (eg.: Usenet groups) simultaneously. As a
verb, the act of posting such spam. The term is derived, obscurely, from a
popular Monty Python comedy sketch that celebrates the pleasures of
consuming a certain tinned meat product.

Thus my reply was:

What spam?? Surely you haven't blocked not only PCDatasheet, but every
other purveyor of their own software, their own agenda, and their own OT
interests ...

While my definition is broader than some, it's valid. When might call this
message spam, actually ...
--
Darryl Kerkeslager

Dec 19 '05 #32

"Darryl Kerkeslager" <ke*********@comcast.net> wrote in
news:MN********************@comcast.com:

What spam?? Surely you haven't blocked not only PCDatasheet, but
every other purveyor of their own software, their own agenda, and
their own OT interests ...

While my definition is broader than some, it's valid. When might
call this message spam, actually ...

It doesn't matter what definition you use. There is no significant
spam problem in this newsgroup. Period.

You have not made any effort to provide even ONE contrary example,
let alone enough to demonstrate that there's some kind of danger of
spam overrunning the newsgroup.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

Dec 20 '05 #33

"David W. Fenton" <XX*******@dfenton.com.invalid> wrote

let alone enough to demonstrate that there's some kind of danger of
spam overrunning the newsgroup.

Never said there was. Now you're just making stuff up.

This is my example, right here. Off-topic crap that nobody else cares
about. This is spam in CDMA.

So I'll stop.
--
Darryl Kerkeslager

Dec 21 '05 #34