Shaun Ginsbourg

A few years ago I cooked beetroot dauphine to go with roast beef for Crhistmas. From memory it worked well. I basically replaced potato plus roasted beetroot and extra flour in a standard potato recipe (I can't remember the proportions). I am guessing that goats cheese might work well in place of diced bacon. Can anyone recommend a tried and tested recipe?

I cook dishes from all around the world and ma po dofu is possibly my favourite of all of them. Made well, it has a most wonderful combination of explosive flavours and delicate textures. Like Luizhou my favourite version (I have tried many) is Fuchsia Dunlop's, with the following notes: 1. I make my own tofu. I have posted my recipe elsewhere on the site. If you are not going to these lengths I would recommend a soft (but not silken) Chinese-style fresh tofu, coagulated with gypsum. Dice it into 2cm cubes. 2. I use Lee Kum Kee brand toban jian as it is easy to get and I believe good quality. The more hardcore option is to find paste which is made in Pixian, which I have managed to do once or twice. It has a more depth of flavour, however its earthy, musty taste may not appeal to everybody. 3. If you use the right tofu you need to take care not to break it up when adding it or completing the dish. Once I have added the tofu I avoid stirring for the remaining stages of the dish and instead shake the pan to coat and mix ingredients. 4. I like to simmer the tofu for a good 8- 10 minutes in the sauce before thickening in order to ensure good penetration of flavour. Be careful not to boil down the sauce too much: there should be plenty of it without it being soupy. 5. I add about 100g peas a minute or two before thickening the sauce. They add a sweetness that balances nicely with the other flavours. If you take this step you can cut back the sugar in Fuschia's recipe to a pinch. I would love to hear other peoples tips for cooking this magnificent dish!

Thanks for your advice Jenni. I have tweaked things a little and managed to get my dosa crepe thin and good colour. For my batter I used a higher rice - lentil ratio than previously. I read that this increases crispiness, although ultimately it is a matter of taste: 500g idli rice 100g parboiled red rice 200g urad gram 50g channa dahl (I read this helps fermentation) 3/4 tsp fenugreek seeds I soaked the lentils and rice separately for 6 hours. I pureed the lentils in the food processor with 1 1/2 cups of water rather than blender until it was a fine but granular paste I ground the rice and fenugreek in the grain mill (see previous post) Finally I added the rice to the food processor and mixed uniformly After 12 hours at 30C nothing much was happening so I added some lukewarm water (perhaps a cups) to make it the consistency of a thick pancake batter, a teaspoon of sugar, and a pinch of dried yeast. I don't know which if any or all of them did the trick, but after a further 12 hours the mix was doubled in volume. At this stage I folded through 1 1/2 tsp salt and enough lukewarm water (about 1 cup) to make it the consistency of thinnish cream and left it a further hour and a half. Obviously you should err on the side of being too thick, as you can always adjust it later. At cooking time I adjusted the consistency by taking a small bowlful out of my pot and adding water or batter to it as needed. I was able to make uttapam with the same batter. This only difference was that I just let it spread naturally. When cooked on one side I sprinkled on my ingredients (raw onions) and these were cooked just enough after I had flipped the cake and cooked it to a good colour on the other side.

One more post whilst I am on a roll ... As well as torillas, I have successfully made excellent tostaditas using my fresh masa. I was mindful of Bayless's warning that home made tortillas generally do not deep fry well because they are too thick and just soak up all the oil. In this limited sense, so he says, factory tortillas are better suited to the purpose. With full size tortillas I have found this to be true. After some experimentation I found that by using smaller quantities of mases (10g instead of the 40g I use for a full size torilla) I get an 8cm disk that comes out of the tortilla press thinner than a full size tortilla and deep fries beautifully, bubbling and puffing up in the middle. They can be done in advance, sprinkled with powdered sea salt and reheated in the oven without losing their crispiness. They make great tostaditas or corn chips. A little labor-intensive, but worth the effort if you are really trying to impress.

I should also note, in response to some previous posts, that Rick Bayless is adamant that neither the nixtamalising corn or the finished product should be refrigerated at any stage and certainly not frozen. He says that fresh masa kept at room temperature, as it should be, perishes within 12 hours. Therefore I start the process 25 - 36 hours before I want tortillas: 24 hours for nixtamalisation and 1 - 12 hour sitting time for the masa. Bayless is also adamant that tortillas can be made a few hours in advance and then reheated (using a steamer and teatowl). However I find that they are best used straight away after they have been cooked and allowed to sit to "steam" briefly whilst wrapped in a tea towel.

I have been making fresh masa for tortillas at home in Melbourne, Australia, for a few weeks now and can post some pictures if there is still any interest given the length of this thread. Although we do not have many Mexican grocers here, I have sourced both yellow and white dried field corn: the former from an African supplier at our local market, and the later from the pet food store. I am using slaked lime sourced from a Thai grocer (similar to that referred to in the excellent blog in the previous post). I have been using Alton Brown's nixtamalisation technique which seems pretty standard. Having read the previous post I will try and refine it and post the results. Grinding, of course, is the biggest challenge. I agree that Alton Brown's suggested use of a food processor for tortilla masa is laughable. My solution is a hand operated Messershmidt steel cone grinder. Grinding with the hand crank was hard work so I motorised the unit using a 1/3 hp motor with a pulley wheel and belt that slows it to about 125 rpm. It is necessary to firmly push the corn down into the grinder but with the motor going this is easy work. The results are splendid. The grinder also works very well for dosa and idli batter. The resulting tortillas have wonderful flavour that beat, hands-down, packaged factory tortillas and even tortillas made freshly from masa harina.

Hi Jenni It's great to hear from a real expert! I have some questions and comments. Questions: 1. I am trying to work out what you mean by urad gram as distinct from urad dal. Is it a different lentil? Or is it just the urad dal whole (but skinned) instead of split? 2. What rice do you use if not basmati? 3. Do you have a favourite potato curry recipe you can share for masala dosa? Preferably one with peas! And how does one make butter masala dosa? This sounds heavenly! 4. Can you share an uttapam recipe? Onion is my favourite. 5. Do you have a poori recipe? After reading your post I googled the Ultra Pride and saw a reference to making poori with it. Prior to this I have been using a shortened flour water dough without the need for grinding. Is this not authentic? Comments: 1. You are quite correct that I meant toor rathen than moong dal. Again Mahdur Jaffrey's version of sambar is delicious but I would love to hear yours (including your recipe for home made sambar powder). 2. Regarding the grinder I originally bought it for grinding nixtamal (damp, treated dried field corn) to make masa for tortillas. When I googled Ultra Pride I read a post by one person who had tried using it to make masa but found it did not grind fine enough. My grinder is adjustable and gives me a good consistency with the rice for dosa. I use a blender for the dahl. The tricky part is combining the two pastes smoothly. It is too thick for the blender so I intend trying the food processor next time which might work given the mixture is already ground. I would be grateful for your suggestions. 3. Jaffrey's coconut chutney recipe is made by adding to the coconut tamarind paste and a ground mixture of urad dal, brown mustard seed and dried chilli fried in a little oil and then mixed with asafatoeda. Can you suggest an alternative?

I have travelled to the south of India several times and thin crispy dosa are one of my fondest culinary memories. I have now succeeded in making them at home using a Madhur Jaffrey recipe. The tricky part is the grinding, for which I now use a motorised grain mill: INGREDIENTS 400g long-grain rice 200g urad dahl salt water vegetable oil METHOD 1. Separately pick over, wash, and soak the rice and lentils in twice their volume of water for two hours. Drain. 2. Separately grind the rice and lentils with a combined total of about 250ml water. Jaffrey says to use one third with the rice and two thirds with the lentils, grinding both (separately) in a blender: the rice to a fine, granular consistency, and the lentils to a smooth, meringue-like consistency. I had no problem grinding the lentils with my 800W Kenwood blender. This was however the blender I bought to replace the previous blender I blew up trying to grind the rice, which was not fine enough in any event. Now I use a motorised steel burr Messerschmidt grain mill for the rice. I have done it by hand with the mill but it was very hard work! By using the mill I don't need to use any extra water and instead put the whole cup with the lentils, making that job easier on my blender. 3. Combine the rice and lentils to as smooth a consistency as possible. I am still working on this part. I think I will have to resort to a food processor for this step next time as the combined paste is two thick for the blender, and you don't want the mix too lumpy. 4. Cover the mixture loosely with plastic wrap and ferment the mixture at 27 - 33 degrees C for about 24 hours. I use a cooler box (in Australia we call it an "Esky", named after the leading brand) in which I have put a 20W lamp and the probe of a thermometer. By partially covering the box with a lid I can maintain a reasonably constant temperature that can be adjusted by moving the lid around. The mixure should at least double in volume. 5. Fold in a cup of lukewarm water and 1 tsp of salt and leave for an hour or so. 6. Heat a large heavy frypan (I find a 16 inch Scanpan is ideal) on medium covered with a thin film of vegetable oil. Ladle about a half cup into the centre and carefuly and gently spread out by moving the back of the ladle in a circular motion until you have a nice thin pancake. To succeed you need batter that is just the right consistency and a deft touch: press too hard and you will put holes in your dosa, press to lightly and it will be too thick. 7. Cook for about 1 1/2 - 2 minutes. Carefully drizzle about a teaspoon of vegetable oil around the edge of the dosa. This will help the edge lift away from the pan. The underside is ready when it is a brown - red colour. 8. As far as I can tell dosas are traditionally cook on only one side. This will become the outside when it is folded over. However until you have achieved a nice thin dosa without lumps in the batter I suggest you turn it over and dook it for 1/2 minute on the other side. 9. Fold it in half and serve it without delay with your favourite condiments. Mine are a shallot and moong dalh curry and coconut chutney. Jaffrey has recipes for these. Potato curry with or without peas are also a classic accompaniment. Why bother, I hear you ask? Like with tortillas made with masa harina, dosa made from packet mix does not compare. Packet dosas use a rising agent that lacks the delicate, sweet-sour flavour that comes from natural fermentation.

In Melbourne Australia we are one month into summer and the abundant wild fennel plants that grow along the banks of creeks are are starting to flower: http://cid-d9bb91bf221f7d67.office.live.com/self.aspx/.Public/eGullet/Wild%20fennel%20flowers.jpg I am keen to try harvesting some pollen but I am not sure whether these flowers are ready for picking. They are quite tiny and seem smaller than many I have seen on the internet. Can anyone advise me whether I should wait?

This recipe will produce a 12 x 19 x ~5cm, ~1kg block of chinese – style tofu that is semi-soft but able to withstand careful stir frying and deep-frying. It is perfect for ma po dofu, which in my view is one of the world's greatest dishes. The technique is adapted from The Book of Tofu, by William Shurtleff and Akiko Aoyagi. I have used calcium sulphate (gypsum) as my coagulant. It is easy to work with and cheap. Magnesium chloride (nigari), which the Japanese use, is supposed to produce better tasting tofu, but I have found it is more difficult to get a good smooth texture because of the speed at which it coagulates. Tofu making is a fickle process. The results will vary according to the quality of the soybeans, coagulant and water used, as well as numerous other factors such as the temperature at which beans are soaked and coagulant added, the length of time the milk is boiled, and the care with which curds are handled. A process that produces excellent tofu one day can fail the next day for no apparent reason. A tofu forming box is highly desireable, if not essential. I built my own wooden box, based on some I saw on the internet. The quantities used in this recipe are designed for my forming box (dimensions below). The box you use should have plenty of holes in the bottom and sides to allow whey to drain away. The box should be nice and deep (high) so that it can hold all the unpressed curds necessary to make a block of decent thickness. Ideally it should have straight rather than tapered sides (as most plastic containers do) so that the lid can cover the entire top surface of tofu block yet lower as the curds are pressed and release liquid. With practice, the following technique will take about 45 minutes time in the kitchen, omitting soaking time for the beans (10 hours or more) and cooling time for the finished tofu (one or two hours), but including washing up, which you can do whilst your pot is on the stove. A soymilk – making machine might reduce this time, although I have not tried it. Sure, you can get decent, fresh tofu from a shop for a couple of bucks without all this trouble. But if you love tofu, you will enjoy the superior flavour of the home-made product. Ingredients 420 g (2 1/3 cups) soybeans 3 ½ l of water 2 ¼ tsp calcium sulphate powder or gypsum Equipment Muslin or cheesecloth, preferably cut into two 50 and 60 cm square pieces 10 l heavy - bottomed pot Forming box 12cm W * 19cm L * 13cm H Large soup ladle or teacup Blender or food processor a 2 l jug and two large bowls kitchen gloves wide paddle or spatula Method Wash soybeans in three changes of water and soak them at room temperature in triple their volume of cold water for at least ten hours. Wash the soaked beans once more then drain. Full the pot with one litre of hot water and set on a hot stove. Working in batches, fill the blender or food processor with beans ¼ - 1/3 full and cover well with water. Puree until you can no longer see bits of beans. Aim to leave ½ - 1 l of water. Transfer the puree to the pot and stir occasionally to prevent burning. After transferring the last batch, use some of the remaining water to rinse the puree left in the processor into the pot. Bring to the boil, stirring. Remove from the heat immediately. Line a bowl with the 60cm square piece of muslin and pour in the bean puree, working in batches if necessary. Wearing kitchen gloves, gather up the edges of the muslin, bring them together, and vigorously squeeze out as much milk into the bowl as possible. Leave the solids inside the muslin. Run cold water over your hands from time to time to relieve them from the heat. Once you have squeezed out all the puree, place the muslin filled with bean solids inside an empty bowl and open out. Pour over the remaining water and leave to sit. Scrub out the pot, fill it with the milk you have already squeezed out, and set it on a medium hot stove. Gather up the edges of the muslin, and squeeze out the remaining milk into the pot. Bring the milk to the boil, stirring to prevent it burning. Simmer for 5 – 7 minutes. Remove from the heat. Leave to cool a few minutes (to 90 degrees Celsius if you have a thermometer). Unless you want yuba to form on top of the milk, you can stir it occasionally to help it cool. Combine the calcium sulphate with about 1/3 cup cold water and stir well to dissolve. Place the paddle in the milk and sweep it back and forth in the milk from one side of the pot to the other. The aim is to get the milk circulating around the pot without splashing it around. With your free hand, give the calcium sulphate mixture a stir and quickly sprinkle it over the surface of the milk. Gradually shorten the sweep of the paddle until bringing it to rest in the middle of the milk. Do not sweep the paddle for more than about 5 seconds after to have added the coagulant. The aim of the exercise is to ensure the coagulant is distributed through the milk without breaking the curds that begin to form once it has been added. Leave the milk undisturbed for about 10 minutes. Carefully line the forming box with the dampened, 50cm square piece of muslin and set it in the sink or a place where the whey can drain out from the box. Using the ladle or tea cup, very carefully scoop the curds that hopefully have formed in the milk and transfer them to the box, filling it evenly. The less broken the curds, the better the texture of the finished product will be. You should see a clear separation between translucent solid curds and clear whey. If not, add more coagulant mixed with water. Do not transfer uncoagulated milk to the forming box, as this will simply drain out the bottom. On the other hand, avoid adding further coagulant unless absolutely necessary – the less coagulant you use, the better the texture of the finished product. Once you have transferred all the curds, carefully fold the excess muslin over the top. Try to distribute the excess muslin evenly so that the top face of the tofu block is a flat as possible. Place the lid on the box and top with weights: about 2 2-1/2 kg is ideal. Leave the weights on the tofu for ½ and hour and then check the firmness of your tofu at 15 or 20 minute intervals. The tofu is ready when whey has stopped draining out of the box or when the firmness is that of well set jelly. Place the box in a bucket or sink full or cold water and carefully remove the muslin block from the box. Leave a few hours before carefully removing the muslin and transferring the block to a sealed contained where it can be covered with water and placed in the fridge. Change the water daily and the tofu will last at least a week.

My experience was that after being dissatisfied with the results I got using a sourdough starter I had "caught" and grown at home (in Melbourne Australia), I mail ordered what was claimed to be the famous San Fransisco culture from the US. I was skeptical about the few flakes that arrived in an envelope, but after faithfully following the instructions I baked successive loaves of wonderful bread. However over the space of a week or two the quality of the bread seemed to deteriorate and became indistinguishable from the bread my home made starter produced. My theory - unsupported by any expertise on the topic - was that the indigenous yeasts took over my lactobacillus sanfranciscensis starter.

Thanks for that Corymoto. So far I have tried 90C for one hour and 85C overnight. Cooking times of 2, 3 hours are not really practical because I don't want to get up really early and if I do it the night before the oats continue to cook and break down overnight anyway. Regarding the use of liquids I have found the difficulties of bagging liquids using a domestic vacuum sealer are overstated by many. I have a foodsaver with a manual "pulse" option that is perfect for things such as porridge. I hold the bottom of the bag over the end of the bench and use gravity to keep liquid away from the vacuum channel. By using the pulse buttom I can remove the air just to the point the liquid starts to move towards the channel and then seal. No problems and no air bubbles. This method works fine for liquid marinates and syrups as well. Theoretically freezing is preferable so that you can get more pressure but I doubt it really makes that much difference.

I can see both pros and cons in using the Sous Vide Supreme instead of a temperature controller / cooker combination. Cons: 1. The combination is potentially cheaper: about $150 for the controller + between $50 and $200 for the cooker depending on size (cf $450 for the Supreme) 2. The combination gives you a potentially bigger water bath. I use a 20 litre urn that cost A$100 on Ebay and could safely cook about 3kg meat - in the 10 litre bath of the Supreme I would guess you are limited to 1 - 2 kg before your heat transfer from water to meat becomes too slow. (Doug Baldwin can probably provide some figures on all this). Pros: 1. Slick design, all-in-one package 2. Less bench space 3. No need to worry about PID settings (although the Sous Vide Magic has an auto-tune function for this) What is not clear from the Sous Vide Supreme specs is: (a) power of the heating element, which determines how quickly you can get your water bath to the right temperature, and (b) how well insulated it is, which determines how efficiently and therefore economically it operates. Also the specs do not indicate there is any circulator built into the device, from which I would assume that the precision of temperature control no better than the combination using cooker that heats from the bottom.

I just had my first (of many I hope) sous vide Xmas dinner. I sourced a rare breed (Wessex Saddleback) suckling pig with a dressed weight of 10-12 kg. My set up consisted of a 20 litre (4 1/2 gal) urn, a Sous-Vide Magic temperature controller, and a 28cm Sunbeam Food Saver. My technique was borrowed from Joan Roca, although I allowed a longer cooking time given his use of a smaller piglet. I broke down the pig and individually bagged the pieces each with a few tablespoons of olive oil and peppercorns. (Olive oil can be solidified in the freezer which helps avoid having to seal liquid using a Food Save). I cooked in three batches, each batch at 70C/158F for 14 hours. Each batch was rapidly cooled in an ice bath and refrigerated in the meat chiller until prior to service. About two hours prior to service I reheated the bags using a domestic oven at 80-85C. After taking the meat out of the bags I removed the bones (after reheating it was possible to remove the bones by just pulling them out). I then attempted to crisp the skin up using non-stick pans on the stove top, using a thermometer to ensure the core temperature was raised enough. The texture of the meat was very good: moist and tender. It was not fork-tender - as I have often seen meat described on this post - but that would not have been to my taste. My efforts to crisp the skin were only partially successful due perhaps to lack of patience and care. But the method was fundamentally sound and I would recommend it. I note that Thomas Keller has a recipe which is basically an assiette that uses a suckling pig of similar dimensions to the one I tackled. It is a highly technical method that involves differential treatment of different parts of the pig. I can see how this would produce better results, even if the results of the simpler technique I used were more than satisfactory. I served the pig with mille-feuille potatoes, shallot confit, sous-vide apple wedges, and a coleslaw made with wombok, fennel, watercress, chervil, and hazelnuts dressed with an apple mustard vinaigrette (Thomas Keller being the inspiration behind some of these accompaniments.) I opted against Joan Roca's accompaniment of orange gastrique which, although delicious-sounding, seemed incompatible with potatoes.

Thanks for your reply Douglas I will give that I try I forgot however to ask one further question: if I am using milk (or 1/2 milk, 1/2 water) as my liquid, what should I consider to be the danger zone in terms of cooking temperature. I would guess that 80C or above would be safe, but again I would appreciate your thoughts.