The rendering of the font in my machine (Windows 7, Chrome 15) is off. You are using your own web font "Cantarell", but it looks faded and unsharpened. Considering that many securities personnel are in offices and a lot of software is Windows-based, this may be an issue for your audience.
Why would this be valuable to traders? Silver is down more than 1 $ today and is still shown as bullish. Am I missing something?The biggest factor this week was option expiry and the fact that there is a massive liquidity squeeze in europe. No one cares what yahoo finance writes. By the time the info is shown on your site is has already been processed by hft algos and been priced in. Sorry for the negative feedback.
- The long list of articles at the bottom of the page seems excessive; why not just show 10 most recent? expand when needed; it just clogs the UI
- Good HTML should read like any other good source code, if the logical, visual, and functional pieces of HTML are broken into disorganized pieces it will be hard for a new member of your team to keep up or to maintain it
- CSS doesn't have any comments [EDIT: this is just for showing your code - if you want good feedback let me understand what you are trying to do - geez]
- You should do some simple clustering or at least use some epsilon in you document similarity matrix to distinguish if an article is new or not; if same or very similar article shows on three different web-sites will it have triple effect on the prices? For example:
- Instead of showing 10 most recent articles it's better to show N-most recent articles with S-most recent trend; this might be difficult to implement but seems more helpful for making a decision; otherwise the user has to do extra work to figure out: "this is an old news, I know that two days ago they thought otherwise, but now there is new trend"
- This is not my domain of expertise and I'm not sure how traders work but this might be useful in healthcare so please keep working on it keeping in mind that this could work much better in different domains
[EDIT:] Try shared neighbor distance method. Whatever you use now for similarity feed it into Jarvis & Patrick method and try different NN. This will take care of outliers and variable densities throughout heterogeneous clusters.
I'm no expert in webdev, but while CSS comments may be useful for maintenance and for new team members, they are also a waste of bandwidth. I'm honestly curious, what HN webdevs do with comments in CSS - do you use them, or not, or maybe have some deployment scripts that strip comments off?
Ideally, you would use a tool like less or sass to generate your css (which we don't), so your comments would be in the underlying file but not in the css given to the user.
Maybe it's useful to support the site, while it's small it's ok, wait 2 more years... We are a very Tech centric community you should expect that kind of comment from HN folks.
Actually, we are doing clustering, but as you found out it's not perfect. Like you said, we don't want the same news repeated over and over again to be factored in the bullbear index.
I like this idea: Instead of showing 10 most recent articles it's better to show N-most recent articles with S-most recent trend.
One more idea. You could add a "system error button" or something similar that a user can click and tell you how your NLP/ML algorithms work. Then, use it for improving your models or maybe even implement active learning.
I am not a lawyer, but I'd be careful to ensure that the advice this application gives does not construe financial advice that would require being registered with a regulator as you could get into serious trouble for that (if you are not registered that is).
A personal preference, but I like to have individual login detail per site. This looks like I can use many other providers to login, but I would prefer to utilize a standard username/password.
I agree with this - please don't make me connect with another login somewhere else. Give the choice of creating a separate username/password for the site.
I had a very similar idea about 15 years ago. I was working for a market research firm that measured the effectiveness of marketing and public relations campaigns. To quantify results we read, categorized and graded articles pertinent to the campaign.
Since we were analyzing thousands of articles a week (this is pre-www BTW) I thought we could leverage that dataset to identify patterns that could be used to predict the market. Unfortunately it turned out to be a lagging indicator and 20/20 hindsight isn't valuable.
I'll be interested to see if you can succeed with essentially the same strategy. Of course the publishing industry is a much different animal now, so maybe it'll work this time around. I'll be watching you closely.
If you are interested in chatting feel free to email me. My address is in my profile.
I agree, predicting purely on lagging information is no good. We had this comment before and that's why we are currently working on distinguishing between a news stating a fact that occurred in the past, present or future. With that, we'll be able to present what the market thinks of the future price movements. For example, an author saying: "we expect copper to surge in the coming months", would be recognized as a future statement and factored in the future bullbear index. Or something like that.
That's smart, and very cool. I recall the dollar signs in my eyes when I thought about the potential in getting a predictive system like this to work -- I hope you guys nail it.
I've learned that it's more important not to trade the news, but trade the reaction to the news.
If a really nasty number comes out (let's say ISM) but the buy programs step in, that tells us that the underlying tape is fairly strong.
Another thing I know about these financial publications is that they are often seeking a reason for market movements. Sometimes it's just pseudo-random price fluctuations.
Also I'm on the main page now, and I"m looking at the sources that you are using. Be very careful sourcing prnewswire.com because, well, they are press releases.
I clicked on a dailymarkets and it was just a syndication of forexpros.com content, try and look for primary source.
I don't know if you're basing readings off the tone of the writing, but consider sourcing in more neutral press from bloomberg/reuters.
Thank you for only requiring the minimum set of permissions from my Facebook account. I hate it when I go to a website that uses Facebook Connect login button, but requires every single permission possible, when I just need an authentication system.
Great tool - nicely done. I have two questions:
1) What is your business model? Hoping for acquisition or hoping to get traders to pay you for data?
2) How lawyered up are you? You're offering financial advice and I know the SEC has some pretty delicate regulations around what you are and are not allowed to charge people for and what the limits of your liability are. Did your legal counsel offer any specific advice in this area before you launched?
Regarding your first question, we want to offer a subscription based service both for the webapp and our API. The API access is not advertised yet, but we are planning to offer an access to our raw data in real-time to be processed by the user.
Yes, well aware of that and we are currently looking into it. Although, we are not pretending to predict the price movement. Rather, we are reporting on what the news is saying.
We are also working on distinguishing between a news stating a fact that occurred in the past, present or future. With that, we'll be able to present what the market thinks of the future price movements.
What's the rationale of not providing a conventional login option? I for one do not have an account with any of listed services - Google, Yahoo, Twitter, etc - and don't have any plans to.
Not a huge deal, but I think you should add the "Try it now" button on your learn more page (http://www.bullbear.ca/more). Clicking on the back button or on the logo to go back to the home page and sign up seems a bit redundant to me.
Also, once you get to the main page, it would be great if by changing your settings you only saw the relevant pieces of news (for example, if I uncheck "Crude Oil" in my settings, I shouldn't have to set a filter on the news as well to remove the elements from the list).
Love the simple interface, very slick. The features I tried worked exactly like I expected them to. Well done!
Interesting idea, and on face value it looks quite useful, however generally this type of trading is only a little better than casino gambling. Unless the algorithm has real valuable insight I would be very dubious as to it's benefits. If it does have really valuable insight then I would expect the creator to use it to run their own fund and investment and post results. Perhaps all that is to come though.
Great tool! I know you probably won't do this, but I think it's important to be clear. How can you convince potential users that you won't front-run them? For example, before displaying the sentiments you run your own trades in anticipation of your customer's trades. Perhaps it'll be right 51% of the time, but that's enough of an edge with enough volume to make some serious money.
Actually I don't think this is illegal. They aren't brokers, or otherwise filling orders, so they can't "front-run" per se. Also, I don't believe front-running is illegal in commodities markets, as they aren't regulated like equity markets.
There was a recent incident where a dark pool provider was frontrunning their own customers. I believe it was Pipeline/Milstream. It happens all the time, and then they get a slap on the wrist.
Thank you all for this feedback. We were amazed by the quality of it.
Note that we have already modified our newsletter subscription process. It is now opt-in instead of opt-out as it should have been from the start. We apologize for that. We have unsubscribed everyone that signed-in since we posted. If anyone wants to hear from us, please subscribe by going to Settings -> Profile.
#1 question to the creators: Do you trade off this information, and if so, what techniques do you use?
I personally can't see how I would be able to trade off this information given the lack of timeliness, so if the creators could talk about how they personally use it, that would make me more interested.
We are looking into adding time information. See me reply to jnorthrop. Essentially, we are working on distinguishing between a news stating a fact that occurred in the past, present or future. Also, a time horizon has been requested before.
Not ourselves, but we know people you do. This whole idea was proposed by a commodity trader friend of ours and we decided to give it a shot. This is what came of it. So far, our trader friend likes it.
I actually work in commodities and I think this is a cracking tool from a quick glance over. I love the extracts you get when you click on a particular news story.
The design could be tweaked a little, but the content is great.
Do you place any emphasis or weighting news sources, or is it all a flat weighting?
Currently the weighting is flat, but that is about to change. What we want to offer, and let me know if that is something of interest to you, is the ability for the user to customize the sources and to customize the weight of each sources. Each traders probably has its own preferred and trusted sources and we want to address that.
Great stuff... I'm working on a similar tool at the moment. Page load speeds were a bit sluggish for me. I'm guessing that's attributable to the large amount of javascript processing happening?
Not only for professional traders but for anyone interested in commodities. We want this tool to become the place to go to get information about the commodity market.
I'd have some doubts about it's utility to a professional trader - they've already got a Bloomberg in front of them, which is telling them everything you can, except faster and from a wider variety of primary sources.
In my experience (which is primarily with options trading, though I've done a stint working on commodity pricing), traders aren't going to rely on a website they know next to nothing about as source of quantifiable information about the markets. Your professional commodities trader is either following their own news sources and following more specialist sources already, and their job (assuming they're not just trading flow or taking client orders) is to turn those sources (along with their own analytics) into sentiment about the NDFs and Future's their trading.
Still, I think it's an interesting idea and definitely one which I'm going to poke around in and see what ideas it can inspire.
What about giving the pros the ability to upload their own documents (and remain private) for BullBear to analyze and report? Also, what about the ability to customize which source of information the system process and which weight each source receive in the bullbear index calculation?
These are all ideas we have on the table for the near future.
If it could do things like that, then maybe selling it as a product that a company can plug their own sources into and let your algos do the crunching and analysis and present the data could be a good move.
As far as we know, they also aggregate news on commodities, but we analyse and aggregate the news by commodity and by identifying price movement information. We also provide a large number of sources, bringing in news from many different providers, let them be new sites, blogs, research papers, etc.
Assuming you want real-time data, this is the sort of thing that would cost a fair amount of money.
Feel free to correct me if I'm wrong. But, from what I can see, this is merely meant as a useful tool for those who are interested in commodities. The people who are seriously interested in trading commodities will probably already have their own favourite price feeds and charting tools.
We use Superfeedr for feeding the news into the python back-end running on AWS. SQS queues are used to send and receive messages between our web portal and analyzers. The analysis is done in real-time and reported to the web application running on Google App Engine. Amazon Simple Notification Service is used to report to the webapp (and will also be used to send to users subscribed to the API). We use S3 to store the raw documents, but CouchDB for the metadata and the results of our algorithms.
Regarding our algorithms, we use Reuters' OpenCalais as a first step to the NLP, but augment it with our own stuff for the bull and bear extraction.
The webapp uses Knockout.js and all other standard stuff like jQuery and underscore.js. We "build" the javascript using require.js.
Why was this posted to hackernews? Most of us aren't commodity traders so can offer nothing more than superficial feedback ( likely on the things traders won't care about - like font choice or css comments...) and we certainly won't be paying customers.
I don't trade. I help traders better understand the market conditions so they can make better trading decisions. Mostly through pulling together different types of publicly available (scraping) and purchased data (or, you could call me a programmer)
After about 5 years of being on the webmaster team of a technology megacorp I was offered a chance to join great energy trading company in a sort of developer/analyst/bridge/free-agent role. Since then, that company has traded hands a couple of times and currently belongs to a large European bank that wanted into the US power market.
I'm creating something myself that whilst it is not commodity trading is based on information processing and prediction. Therefore I find this very interesting. It's also interesting in terms of what technologies were used (which has been answered in a another part of the comments). I actually think this is a good fit and its also someone sharing with the community what they have built. Good for them
