seulater:
I now am beginning to think it was more a SF scam to test their servers in some way. I say SF should post a list of all the orders made.
And their addresses so we can beat them up. :lol:
And their addresses so we can beat them up.
YA! 
Come on guys, its not like Woot.com 's Bag of Crap that sells out in milliseconds… And they really did try to beef up the servers.
None of my friends got anything, but they got close.
They should have set the server up to give each user a specific amount of bandwidth, until all the bandwidth was used up. At that point the server would not give out any more ports for new connections until the some of the previous connections have completed their orders and left. To just simply hand out ports until all ports are used up and no one has any usable bandwidth just makes no sense.
Even if the above is not correct. The closer you are to SFE the better your chances are for getting in.
I say IF SFE ever even thinks of doing this again, they should make sure EVERYONE has a fair shot to get in.
rbz:
I think it was just people in the same state who got served.
I’m in a different timezone than Sparkfun.
seulater:
Come on guys, its not like Woot.com 's Bag of Crap that sells out in milliseconds… And they really did try to beef up the servers.
None of my friends got anything, but they got close.
They should have set the server up to give each user a specific amount of bandwidth, until all the bandwidth was used up. At that point the server would not give out any more ports for new connections until the some of the previous connections have completed their orders and left. To just simply hand out ports until all ports are used up and no one has any usable bandwidth just makes no sense.
Even if the above is not correct. The closer you are to SFE the better your chances are for getting in.
I say IF SFE ever even thinks of doing this again, they should make sure EVERYONE has a fair shot to get in.
Do you have any idea of what you are saying? You want them to rewrite a network stack to fit your idea of “fair”? Proximity has nothing to do with timing and internet topologies.
Do you have any idea of what you are saying? You want them to rewrite a network stack to fit your idea of “fair”?
I would imagine so, if my tiny tiny network server can do this with a couple clicks, i would certainly hope that SF MUCH larger one could do the same and more, its not difficult at all to set it up that way. The basic two configs are release ports are they are requested, or throttle the bandwidth to each port until bandwidth is used up.
Proximity has nothing to do with timing and internet topologies.
Are you kidding me ?!?!
a ping from 1 mile away is going to be less that a ping for 2000 miles away, of course it maters. There are some exceptions to this according to routing WAN networks.
Monday morning quarterbacks are all over the place.
I spent 2 hours getting time out errors. I’ll never get those back. I’m not mad at SFE. However, I am mad at myself for believing their bull about “we’ve really beefed up our servers”. Bah, it was amateur hour. Sour taste in my mouth - for sure.
Nate, if you read this think about the magnification factor that applies to angry customers. They tell 10 times the number of people about their experience that happy customers do. If you ever do this again, figure out some way to avoid “router roulette”. The impression is that the script kiddies gamed the system and loyal customers lost out.
If you ever do this again, figure out some way to avoid “router roulette”. The impression is that the script kiddies gamed the system and loyal customers lost out.
Well said… Well Said…
seulater:
I would imagine so, if my tiny tiny network server can do this with a couple clicks, i would certainly hope that SF MUCH larger one could do the same and more, its not difficult at all to set it up that way. The basic two configs are release ports are they are requested, or throttle the bandwidth to each port until bandwidth is used up.
All the bandwidth, ports, and throttling in the world can’t fix bottlenecked code/database interactions…and rewriting a network stack entails far more than a few clicks. Familiar with Linux TCP stacks?
Are you kidding me ?!?!
a ping from 1 mile away is going to be less that a ping for 2000 miles away, of course it maters. There are some exceptions to this according to routing WAN networks.
Which is precisely why I mentioned “internet topologies”. Routing through a slow network will slow you down, even if you are just a mile away. This is simple networking principles here.
My point with all of this is that there are so many factors involved, that a simple blanket statement on what is “fair” a moot point.
FWIW, I think the anger is a bit misdirected. I get irritated when I’m trying to fight the rest of the world for concert tickets on TicketMaster and their site goes down–they do it for a living, know exactly what to expect, and still can’t handle the load. SFE’s server doesn’t seem good at handling “normal” daily load, so it’s hard to be surprised that it couldn’t handle the surge.
I laughed out loud when my page refreshed itself right at the start. Probably not the best idea…
I got my order through at about the half-way mark with a lot of hitting “Try Again” every five minutes. I wasn’t going to post in a complaint thread since I don’t have the right to complain about anything, but I guess it’s worth knowing that there are some real repeat-customers that did get through. I don’t know how many script kiddies were trying to get free stuff, but I think the result would have been just as bad if all the users were legit.
And no, you can’t have my address
If SF holds another event of this magnitude, I would suggest reworking the checkout page to make it more possible to get an order through even when the server is under load.
Specifically, the problem I had was, once you get to the https://www.sparkfun.com/commerce/checkout_payment.php page, the “Continue” button sends you right back to the same page. I assume it would normally show an order confirmation screen, but if you lose connection going to that page and refresh, you’re stuck at entering your payment info again. Then, you click continue, lose connection, refresh a few times, and start all over again. When you’ve gone through this loop repeatedly over the course of a couple hours, it gets a little frustrating knowing that you’re so close, yet so far, from making your order. I don’t claim to know how this could be fixed, but if this was set up so that losing connection didn’t mean starting over, you’d still be able to get an order through even if your connection is intermittent.
Philba:
The impression is that the script kiddies gamed the system and loyal customers lost out.
But honestly, did you think it would be any other way? From the start I expected that the site would be effectively Slashdotted and that only college & high school kids would have the patience to stick around trying to order.
I tried for a short time, gave up, and forgot about it. Sure there was some stuff I would have liked, but it’s not a big deal. If I really want it, then I’ll save my pennies and pay for it. Free stuff is nice, but I think it’s pretty bad form to complain if I don’t get any!
In other news: Life’s Not Fair.
Hardly. Their cost of goods will be much lower than $100K considering their normal markup and the fact that the winners still had to pay shipping costs.noctolater:
the ‘free’ publicity cost them $100,000, remember
Hope springs eternal - I went pretty much the full event pushing retry/refresh hoping to get in. Like I said earlier, I’m not mad at SFE but do have a diminished regard for them. or perhaps, it’s a more realistic regard.
Regardless of the arguments put forth, it’s clear that Free Day did not have a positive effect on the majority of customers. This is not good for SFE no matter whether it’s just a bunch of whiners or what not. They are all potential customers that have a negative reaction running from mildly disappointed to pissed off. What appeared brilliant a week ago now appears ill thought out. (many say bone-headed)
SodaAnt:
Hardly. Their cost of goods will be much lower than $100K considering their normal markup and the fact that the winners still had to pay shipping costs.noctolater:
the ‘free’ publicity cost them $100,000, remember
You are forgetting opportunity cost. They lost a day’s worth of sales and spent a lot of time preparing plus the massive headache of shipping 1000 packages in a reasonable timeframe. I think the total bill is probably more than $100K.
And given the angst that is being shown in the forums, there is a fair amount of negative intangible as well.
So yeah, i am a returning customer, and i spent 3 days getting my basket prepped for this, and i get to the checkout button, spend 20 minutes on waiting for the servers to respond, and then the free day is done!!1
THIS
``` IS BS AND I TINK WE SHOULD BE ABLE TO RECEIVE SOME COMPENSATION FOR THIS!!!!!!!!!!!!!!!!!!!!!!!! :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :!: :!: :!: :!: :evil: :evil: :evil:led235:
So yeah, i am a returning customer, and i spent 3 days getting my basket prepped for this, and i get to the checkout button, spend 20 minutes on waiting for the servers to respond, and then the free day is done!!1THIS ``` IS BS AND I TINK WE SHOULD BE ABLE TO RECEIVE SOME COMPENSATION FOR THIS!!!!!!!!!!!!!!!!!!!!!!!! :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :evil: :!: :!: :!: :!: :evil: :evil: :evil:
I think you should compensate me for reading your post.
I want to see someone getting fired over this, who in their right mind mixes a server test with a marketing day, whatever you wanted to do you did it wrong.
qwertyuiop:
I want to see someone getting fired over this, who in their right mind mixes a server test with a marketing day, whatever you wanted to do you did it wrong.
Your panties are in a serious bunch. You need to relax and lay off the conspiracy theories.
-Bill
I am %100 convinced that this was a total scam by Sparkfun.
What a better way to get free server testing, extensive viral marketing and attention. The best part is that we have no way to verify if they actually gave away $100k worth of stuff since they are a private company. All they had to do was allow a few random people the deal so that at least some people could verify that it went through.
In the end, most people who know about and would use Sparkfun already were customers and are now pissed off at having their time wasted by a dangling carrot. The majority of traffic was likely from people who simply could not resist the idea of free stuff (and probably would have got things they would never use).
I recommend going to other places like seeedstudio. Way better customer service and often better prices.